Model save
Browse files- README.md +66 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep13_04-33-54_COE-CS-sv003/events.out.tfevents.1726202353.COE-CS-sv003.3184711.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2787 -0
README.md
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: feedback_p0.1_seed42_level2_syntaxmixbatch16
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# feedback_p0.1_seed42_level2_syntaxmixbatch16
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.3189
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 4
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- total_train_batch_size: 16
|
| 48 |
+
- total_eval_batch_size: 4
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 1
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 58 |
+
| 0.3656 | 1.0 | 1950 | 0.3189 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
### Framework versions
|
| 62 |
+
|
| 63 |
+
- Transformers 4.43.4
|
| 64 |
+
- Pytorch 2.3.1+cu121
|
| 65 |
+
- Datasets 2.19.1
|
| 66 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 204145164288000.0,
|
| 4 |
+
"train_loss": 0.5252774189068721,
|
| 5 |
+
"train_runtime": 19779.9981,
|
| 6 |
+
"train_samples": 98927,
|
| 7 |
+
"train_samples_per_second": 1.577,
|
| 8 |
+
"train_steps_per_second": 0.099
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.43.4"
|
| 6 |
+
}
|
runs/Sep13_04-33-54_COE-CS-sv003/events.out.tfevents.1726202353.COE-CS-sv003.3184711.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:984ef5d2e8fbb867b178b50ad284ddab8f76d8054a8a326984e86de1a5a08b30
|
| 3 |
+
size 88290
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 204145164288000.0,
|
| 4 |
+
"train_loss": 0.5252774189068721,
|
| 5 |
+
"train_runtime": 19779.9981,
|
| 6 |
+
"train_samples": 98927,
|
| 7 |
+
"train_samples_per_second": 1.577,
|
| 8 |
+
"train_steps_per_second": 0.099
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2787 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1950,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005128205128205128,
|
| 13 |
+
"grad_norm": 27.855213353292697,
|
| 14 |
+
"learning_rate": 5.1282051282051286e-08,
|
| 15 |
+
"loss": 1.4714,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002564102564102564,
|
| 20 |
+
"grad_norm": 24.297444748053664,
|
| 21 |
+
"learning_rate": 2.564102564102564e-07,
|
| 22 |
+
"loss": 1.3999,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005128205128205128,
|
| 27 |
+
"grad_norm": 16.726484469297354,
|
| 28 |
+
"learning_rate": 5.128205128205128e-07,
|
| 29 |
+
"loss": 1.3624,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007692307692307693,
|
| 34 |
+
"grad_norm": 12.476091472369891,
|
| 35 |
+
"learning_rate": 7.692307692307694e-07,
|
| 36 |
+
"loss": 1.2347,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010256410256410256,
|
| 41 |
+
"grad_norm": 8.35097021976234,
|
| 42 |
+
"learning_rate": 1.0256410256410257e-06,
|
| 43 |
+
"loss": 1.121,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01282051282051282,
|
| 48 |
+
"grad_norm": 3.6024304396596554,
|
| 49 |
+
"learning_rate": 1.282051282051282e-06,
|
| 50 |
+
"loss": 0.9839,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015384615384615385,
|
| 55 |
+
"grad_norm": 3.480763645073715,
|
| 56 |
+
"learning_rate": 1.5384615384615387e-06,
|
| 57 |
+
"loss": 0.9315,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.017948717948717947,
|
| 62 |
+
"grad_norm": 3.0807184674746546,
|
| 63 |
+
"learning_rate": 1.794871794871795e-06,
|
| 64 |
+
"loss": 0.9081,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.020512820512820513,
|
| 69 |
+
"grad_norm": 2.915980172201366,
|
| 70 |
+
"learning_rate": 2.0512820512820513e-06,
|
| 71 |
+
"loss": 0.8695,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.023076923076923078,
|
| 76 |
+
"grad_norm": 2.9407435214657136,
|
| 77 |
+
"learning_rate": 2.307692307692308e-06,
|
| 78 |
+
"loss": 0.8613,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02564102564102564,
|
| 83 |
+
"grad_norm": 2.9974488091505522,
|
| 84 |
+
"learning_rate": 2.564102564102564e-06,
|
| 85 |
+
"loss": 0.8461,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.028205128205128206,
|
| 90 |
+
"grad_norm": 3.2196932971809713,
|
| 91 |
+
"learning_rate": 2.8205128205128207e-06,
|
| 92 |
+
"loss": 0.8273,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03076923076923077,
|
| 97 |
+
"grad_norm": 3.148550996118887,
|
| 98 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 99 |
+
"loss": 0.8436,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03333333333333333,
|
| 104 |
+
"grad_norm": 3.220886247551075,
|
| 105 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 106 |
+
"loss": 0.8211,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.035897435897435895,
|
| 111 |
+
"grad_norm": 3.0507943848526597,
|
| 112 |
+
"learning_rate": 3.58974358974359e-06,
|
| 113 |
+
"loss": 0.8178,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.038461538461538464,
|
| 118 |
+
"grad_norm": 2.9585065948581484,
|
| 119 |
+
"learning_rate": 3.846153846153847e-06,
|
| 120 |
+
"loss": 0.8077,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.041025641025641026,
|
| 125 |
+
"grad_norm": 3.0972764668895576,
|
| 126 |
+
"learning_rate": 4.102564102564103e-06,
|
| 127 |
+
"loss": 0.7975,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04358974358974359,
|
| 132 |
+
"grad_norm": 3.042612736519298,
|
| 133 |
+
"learning_rate": 4.358974358974359e-06,
|
| 134 |
+
"loss": 0.802,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.046153846153846156,
|
| 139 |
+
"grad_norm": 3.249820635668479,
|
| 140 |
+
"learning_rate": 4.615384615384616e-06,
|
| 141 |
+
"loss": 0.7839,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.04871794871794872,
|
| 146 |
+
"grad_norm": 2.9967690432516747,
|
| 147 |
+
"learning_rate": 4.871794871794872e-06,
|
| 148 |
+
"loss": 0.7686,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05128205128205128,
|
| 153 |
+
"grad_norm": 3.3554959103069737,
|
| 154 |
+
"learning_rate": 5.128205128205128e-06,
|
| 155 |
+
"loss": 0.7628,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05384615384615385,
|
| 160 |
+
"grad_norm": 3.0970153076575975,
|
| 161 |
+
"learning_rate": 5.384615384615385e-06,
|
| 162 |
+
"loss": 0.7564,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05641025641025641,
|
| 167 |
+
"grad_norm": 3.19251255800907,
|
| 168 |
+
"learning_rate": 5.641025641025641e-06,
|
| 169 |
+
"loss": 0.7521,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05897435897435897,
|
| 174 |
+
"grad_norm": 2.989469084400876,
|
| 175 |
+
"learning_rate": 5.897435897435898e-06,
|
| 176 |
+
"loss": 0.7497,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06153846153846154,
|
| 181 |
+
"grad_norm": 2.9701683904826743,
|
| 182 |
+
"learning_rate": 6.153846153846155e-06,
|
| 183 |
+
"loss": 0.7482,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0641025641025641,
|
| 188 |
+
"grad_norm": 2.806869258124627,
|
| 189 |
+
"learning_rate": 6.410256410256412e-06,
|
| 190 |
+
"loss": 0.7584,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06666666666666667,
|
| 195 |
+
"grad_norm": 2.951227277607944,
|
| 196 |
+
"learning_rate": 6.666666666666667e-06,
|
| 197 |
+
"loss": 0.744,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.06923076923076923,
|
| 202 |
+
"grad_norm": 3.0730318127191385,
|
| 203 |
+
"learning_rate": 6.923076923076923e-06,
|
| 204 |
+
"loss": 0.7413,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07179487179487179,
|
| 209 |
+
"grad_norm": 2.8953064927208025,
|
| 210 |
+
"learning_rate": 7.17948717948718e-06,
|
| 211 |
+
"loss": 0.738,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07435897435897436,
|
| 216 |
+
"grad_norm": 2.9241997829974147,
|
| 217 |
+
"learning_rate": 7.435897435897437e-06,
|
| 218 |
+
"loss": 0.7274,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07692307692307693,
|
| 223 |
+
"grad_norm": 2.885045926032792,
|
| 224 |
+
"learning_rate": 7.692307692307694e-06,
|
| 225 |
+
"loss": 0.7296,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.07948717948717948,
|
| 230 |
+
"grad_norm": 3.0783343062947184,
|
| 231 |
+
"learning_rate": 7.948717948717949e-06,
|
| 232 |
+
"loss": 0.7154,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08205128205128205,
|
| 237 |
+
"grad_norm": 2.9470190275773542,
|
| 238 |
+
"learning_rate": 8.205128205128205e-06,
|
| 239 |
+
"loss": 0.7215,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08461538461538462,
|
| 244 |
+
"grad_norm": 3.28788682771588,
|
| 245 |
+
"learning_rate": 8.461538461538462e-06,
|
| 246 |
+
"loss": 0.7314,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08717948717948718,
|
| 251 |
+
"grad_norm": 2.839890562503017,
|
| 252 |
+
"learning_rate": 8.717948717948719e-06,
|
| 253 |
+
"loss": 0.7119,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.08974358974358974,
|
| 258 |
+
"grad_norm": 2.908179397354329,
|
| 259 |
+
"learning_rate": 8.974358974358976e-06,
|
| 260 |
+
"loss": 0.7201,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09230769230769231,
|
| 265 |
+
"grad_norm": 2.775686236070062,
|
| 266 |
+
"learning_rate": 9.230769230769232e-06,
|
| 267 |
+
"loss": 0.732,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09487179487179487,
|
| 272 |
+
"grad_norm": 2.8091812680053803,
|
| 273 |
+
"learning_rate": 9.487179487179487e-06,
|
| 274 |
+
"loss": 0.7192,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.09743589743589744,
|
| 279 |
+
"grad_norm": 3.043641292876537,
|
| 280 |
+
"learning_rate": 9.743589743589744e-06,
|
| 281 |
+
"loss": 0.7289,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1,
|
| 286 |
+
"grad_norm": 2.806409321185608,
|
| 287 |
+
"learning_rate": 1e-05,
|
| 288 |
+
"loss": 0.7211,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10256410256410256,
|
| 293 |
+
"grad_norm": 2.71943069020764,
|
| 294 |
+
"learning_rate": 9.999799726899261e-06,
|
| 295 |
+
"loss": 0.7212,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10512820512820513,
|
| 300 |
+
"grad_norm": 2.783096760671244,
|
| 301 |
+
"learning_rate": 9.999198923640774e-06,
|
| 302 |
+
"loss": 0.7203,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1076923076923077,
|
| 307 |
+
"grad_norm": 2.9232797812449123,
|
| 308 |
+
"learning_rate": 9.998197638354428e-06,
|
| 309 |
+
"loss": 0.7247,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11025641025641025,
|
| 314 |
+
"grad_norm": 2.6455216659425345,
|
| 315 |
+
"learning_rate": 9.996795951252427e-06,
|
| 316 |
+
"loss": 0.7047,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11282051282051282,
|
| 321 |
+
"grad_norm": 2.770746104512415,
|
| 322 |
+
"learning_rate": 9.994993974622863e-06,
|
| 323 |
+
"loss": 0.7237,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11538461538461539,
|
| 328 |
+
"grad_norm": 2.942090186870347,
|
| 329 |
+
"learning_rate": 9.992791852820709e-06,
|
| 330 |
+
"loss": 0.7318,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.11794871794871795,
|
| 335 |
+
"grad_norm": 2.8360683350136324,
|
| 336 |
+
"learning_rate": 9.990189762256275e-06,
|
| 337 |
+
"loss": 0.7223,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12051282051282051,
|
| 342 |
+
"grad_norm": 2.7028129721998035,
|
| 343 |
+
"learning_rate": 9.987187911381059e-06,
|
| 344 |
+
"loss": 0.6966,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12307692307692308,
|
| 349 |
+
"grad_norm": 2.8576067757026955,
|
| 350 |
+
"learning_rate": 9.983786540671052e-06,
|
| 351 |
+
"loss": 0.6921,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.12564102564102564,
|
| 356 |
+
"grad_norm": 3.1181897606257025,
|
| 357 |
+
"learning_rate": 9.979985922607476e-06,
|
| 358 |
+
"loss": 0.6892,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1282051282051282,
|
| 363 |
+
"grad_norm": 2.6050383237898274,
|
| 364 |
+
"learning_rate": 9.975786361654959e-06,
|
| 365 |
+
"loss": 0.6907,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13076923076923078,
|
| 370 |
+
"grad_norm": 2.588587494074759,
|
| 371 |
+
"learning_rate": 9.971188194237141e-06,
|
| 372 |
+
"loss": 0.7072,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13333333333333333,
|
| 377 |
+
"grad_norm": 2.583793235527538,
|
| 378 |
+
"learning_rate": 9.966191788709716e-06,
|
| 379 |
+
"loss": 0.7059,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1358974358974359,
|
| 384 |
+
"grad_norm": 2.585476026019637,
|
| 385 |
+
"learning_rate": 9.960797545330936e-06,
|
| 386 |
+
"loss": 0.7125,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.13846153846153847,
|
| 391 |
+
"grad_norm": 2.7111790603577877,
|
| 392 |
+
"learning_rate": 9.955005896229543e-06,
|
| 393 |
+
"loss": 0.6997,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14102564102564102,
|
| 398 |
+
"grad_norm": 2.546803585442374,
|
| 399 |
+
"learning_rate": 9.948817305370145e-06,
|
| 400 |
+
"loss": 0.6886,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14358974358974358,
|
| 405 |
+
"grad_norm": 2.504186693121665,
|
| 406 |
+
"learning_rate": 9.942232268516051e-06,
|
| 407 |
+
"loss": 0.6904,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.14615384615384616,
|
| 412 |
+
"grad_norm": 2.423059673415451,
|
| 413 |
+
"learning_rate": 9.935251313189564e-06,
|
| 414 |
+
"loss": 0.7086,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.14871794871794872,
|
| 419 |
+
"grad_norm": 2.5596326656143806,
|
| 420 |
+
"learning_rate": 9.927874998629714e-06,
|
| 421 |
+
"loss": 0.6893,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15128205128205127,
|
| 426 |
+
"grad_norm": 2.658800164746175,
|
| 427 |
+
"learning_rate": 9.920103915747452e-06,
|
| 428 |
+
"loss": 0.6819,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15384615384615385,
|
| 433 |
+
"grad_norm": 2.5618934622112106,
|
| 434 |
+
"learning_rate": 9.911938687078324e-06,
|
| 435 |
+
"loss": 0.6823,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.1564102564102564,
|
| 440 |
+
"grad_norm": 2.5974964155912903,
|
| 441 |
+
"learning_rate": 9.9033799667326e-06,
|
| 442 |
+
"loss": 0.7085,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.15897435897435896,
|
| 447 |
+
"grad_norm": 2.607874540525578,
|
| 448 |
+
"learning_rate": 9.89442844034286e-06,
|
| 449 |
+
"loss": 0.6766,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16153846153846155,
|
| 454 |
+
"grad_norm": 2.732487640942877,
|
| 455 |
+
"learning_rate": 9.885084825009085e-06,
|
| 456 |
+
"loss": 0.6863,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1641025641025641,
|
| 461 |
+
"grad_norm": 2.5827249383659865,
|
| 462 |
+
"learning_rate": 9.875349869241202e-06,
|
| 463 |
+
"loss": 0.6994,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.16666666666666666,
|
| 468 |
+
"grad_norm": 2.5074970797423,
|
| 469 |
+
"learning_rate": 9.86522435289912e-06,
|
| 470 |
+
"loss": 0.672,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.16923076923076924,
|
| 475 |
+
"grad_norm": 2.535326853580366,
|
| 476 |
+
"learning_rate": 9.854709087130261e-06,
|
| 477 |
+
"loss": 0.6842,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.1717948717948718,
|
| 482 |
+
"grad_norm": 2.5728898533459663,
|
| 483 |
+
"learning_rate": 9.843804914304578e-06,
|
| 484 |
+
"loss": 0.6826,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17435897435897435,
|
| 489 |
+
"grad_norm": 2.3552532454072503,
|
| 490 |
+
"learning_rate": 9.83251270794707e-06,
|
| 491 |
+
"loss": 0.6947,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.17692307692307693,
|
| 496 |
+
"grad_norm": 2.37727229081133,
|
| 497 |
+
"learning_rate": 9.820833372667813e-06,
|
| 498 |
+
"loss": 0.676,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1794871794871795,
|
| 503 |
+
"grad_norm": 2.4920852746047966,
|
| 504 |
+
"learning_rate": 9.80876784408948e-06,
|
| 505 |
+
"loss": 0.678,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18205128205128204,
|
| 510 |
+
"grad_norm": 2.6448068047121156,
|
| 511 |
+
"learning_rate": 9.796317088772402e-06,
|
| 512 |
+
"loss": 0.6708,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.18461538461538463,
|
| 517 |
+
"grad_norm": 2.6210733196063174,
|
| 518 |
+
"learning_rate": 9.783482104137127e-06,
|
| 519 |
+
"loss": 0.6641,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.18717948717948718,
|
| 524 |
+
"grad_norm": 2.4774458867373337,
|
| 525 |
+
"learning_rate": 9.770263918384523e-06,
|
| 526 |
+
"loss": 0.6678,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.18974358974358974,
|
| 531 |
+
"grad_norm": 2.4109738214433567,
|
| 532 |
+
"learning_rate": 9.75666359041341e-06,
|
| 533 |
+
"loss": 0.6664,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19230769230769232,
|
| 538 |
+
"grad_norm": 2.6223339803254118,
|
| 539 |
+
"learning_rate": 9.742682209735727e-06,
|
| 540 |
+
"loss": 0.6865,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.19487179487179487,
|
| 545 |
+
"grad_norm": 2.5414701027854787,
|
| 546 |
+
"learning_rate": 9.728320896389263e-06,
|
| 547 |
+
"loss": 0.6951,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.19743589743589743,
|
| 552 |
+
"grad_norm": 2.379668261870325,
|
| 553 |
+
"learning_rate": 9.713580800847917e-06,
|
| 554 |
+
"loss": 0.6672,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2,
|
| 559 |
+
"grad_norm": 2.444662137369538,
|
| 560 |
+
"learning_rate": 9.698463103929542e-06,
|
| 561 |
+
"loss": 0.6622,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.20256410256410257,
|
| 566 |
+
"grad_norm": 2.449846598434235,
|
| 567 |
+
"learning_rate": 9.682969016701357e-06,
|
| 568 |
+
"loss": 0.65,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.20512820512820512,
|
| 573 |
+
"grad_norm": 2.4410258623541763,
|
| 574 |
+
"learning_rate": 9.66709978038292e-06,
|
| 575 |
+
"loss": 0.6608,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2076923076923077,
|
| 580 |
+
"grad_norm": 2.3782584039118015,
|
| 581 |
+
"learning_rate": 9.650856666246693e-06,
|
| 582 |
+
"loss": 0.6566,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.21025641025641026,
|
| 587 |
+
"grad_norm": 2.4939928679664773,
|
| 588 |
+
"learning_rate": 9.63424097551621e-06,
|
| 589 |
+
"loss": 0.645,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2128205128205128,
|
| 594 |
+
"grad_norm": 2.431913499962849,
|
| 595 |
+
"learning_rate": 9.617254039261835e-06,
|
| 596 |
+
"loss": 0.6333,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2153846153846154,
|
| 601 |
+
"grad_norm": 2.4550799694984318,
|
| 602 |
+
"learning_rate": 9.599897218294122e-06,
|
| 603 |
+
"loss": 0.6716,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.21794871794871795,
|
| 608 |
+
"grad_norm": 2.454256850201106,
|
| 609 |
+
"learning_rate": 9.582171903054815e-06,
|
| 610 |
+
"loss": 0.643,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2205128205128205,
|
| 615 |
+
"grad_norm": 2.38325690019972,
|
| 616 |
+
"learning_rate": 9.564079513505455e-06,
|
| 617 |
+
"loss": 0.6351,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2230769230769231,
|
| 622 |
+
"grad_norm": 2.4210446447570533,
|
| 623 |
+
"learning_rate": 9.54562149901362e-06,
|
| 624 |
+
"loss": 0.6627,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.22564102564102564,
|
| 629 |
+
"grad_norm": 2.479279392696381,
|
| 630 |
+
"learning_rate": 9.526799338236828e-06,
|
| 631 |
+
"loss": 0.6506,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2282051282051282,
|
| 636 |
+
"grad_norm": 2.577651511733567,
|
| 637 |
+
"learning_rate": 9.507614539004082e-06,
|
| 638 |
+
"loss": 0.6306,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.23076923076923078,
|
| 643 |
+
"grad_norm": 2.416794598688409,
|
| 644 |
+
"learning_rate": 9.488068638195072e-06,
|
| 645 |
+
"loss": 0.6401,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.23333333333333334,
|
| 650 |
+
"grad_norm": 2.746257367697141,
|
| 651 |
+
"learning_rate": 9.468163201617063e-06,
|
| 652 |
+
"loss": 0.6299,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2358974358974359,
|
| 657 |
+
"grad_norm": 2.443879358348906,
|
| 658 |
+
"learning_rate": 9.447899823879456e-06,
|
| 659 |
+
"loss": 0.6453,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.23846153846153847,
|
| 664 |
+
"grad_norm": 2.3287172810254773,
|
| 665 |
+
"learning_rate": 9.427280128266049e-06,
|
| 666 |
+
"loss": 0.628,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.24102564102564103,
|
| 671 |
+
"grad_norm": 2.462832532756568,
|
| 672 |
+
"learning_rate": 9.406305766604996e-06,
|
| 673 |
+
"loss": 0.6266,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.24358974358974358,
|
| 678 |
+
"grad_norm": 2.6197702096400937,
|
| 679 |
+
"learning_rate": 9.384978419136469e-06,
|
| 680 |
+
"loss": 0.632,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.24615384615384617,
|
| 685 |
+
"grad_norm": 2.457574610754138,
|
| 686 |
+
"learning_rate": 9.363299794378072e-06,
|
| 687 |
+
"loss": 0.6218,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.24871794871794872,
|
| 692 |
+
"grad_norm": 2.3841136661317552,
|
| 693 |
+
"learning_rate": 9.34127162898797e-06,
|
| 694 |
+
"loss": 0.6192,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2512820512820513,
|
| 699 |
+
"grad_norm": 2.3066096761948383,
|
| 700 |
+
"learning_rate": 9.318895687625752e-06,
|
| 701 |
+
"loss": 0.6285,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.25384615384615383,
|
| 706 |
+
"grad_norm": 2.4121900484386805,
|
| 707 |
+
"learning_rate": 9.296173762811084e-06,
|
| 708 |
+
"loss": 0.6214,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2564102564102564,
|
| 713 |
+
"grad_norm": 2.401649957120604,
|
| 714 |
+
"learning_rate": 9.273107674780102e-06,
|
| 715 |
+
"loss": 0.6295,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.258974358974359,
|
| 720 |
+
"grad_norm": 2.3261987507248825,
|
| 721 |
+
"learning_rate": 9.249699271339594e-06,
|
| 722 |
+
"loss": 0.6165,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.26153846153846155,
|
| 727 |
+
"grad_norm": 2.4189305418813336,
|
| 728 |
+
"learning_rate": 9.225950427718974e-06,
|
| 729 |
+
"loss": 0.6148,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2641025641025641,
|
| 734 |
+
"grad_norm": 2.3669436725839543,
|
| 735 |
+
"learning_rate": 9.201863046420065e-06,
|
| 736 |
+
"loss": 0.6374,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.26666666666666666,
|
| 741 |
+
"grad_norm": 2.323600220845717,
|
| 742 |
+
"learning_rate": 9.177439057064684e-06,
|
| 743 |
+
"loss": 0.6168,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.2692307692307692,
|
| 748 |
+
"grad_norm": 2.48316209601843,
|
| 749 |
+
"learning_rate": 9.152680416240059e-06,
|
| 750 |
+
"loss": 0.6163,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2717948717948718,
|
| 755 |
+
"grad_norm": 2.491614167030535,
|
| 756 |
+
"learning_rate": 9.1275891073421e-06,
|
| 757 |
+
"loss": 0.61,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2743589743589744,
|
| 762 |
+
"grad_norm": 2.644322477647823,
|
| 763 |
+
"learning_rate": 9.102167140416503e-06,
|
| 764 |
+
"loss": 0.6274,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.27692307692307694,
|
| 769 |
+
"grad_norm": 2.3420947737074957,
|
| 770 |
+
"learning_rate": 9.076416551997721e-06,
|
| 771 |
+
"loss": 0.5981,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2794871794871795,
|
| 776 |
+
"grad_norm": 2.4856053771378086,
|
| 777 |
+
"learning_rate": 9.050339404945834e-06,
|
| 778 |
+
"loss": 0.6134,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.28205128205128205,
|
| 783 |
+
"grad_norm": 2.3446213277401995,
|
| 784 |
+
"learning_rate": 9.023937788281278e-06,
|
| 785 |
+
"loss": 0.6014,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.2846153846153846,
|
| 790 |
+
"grad_norm": 3.401791255832871,
|
| 791 |
+
"learning_rate": 8.997213817017508e-06,
|
| 792 |
+
"loss": 0.5976,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.28717948717948716,
|
| 797 |
+
"grad_norm": 2.511130997702265,
|
| 798 |
+
"learning_rate": 8.970169631991556e-06,
|
| 799 |
+
"loss": 0.5913,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.28974358974358977,
|
| 804 |
+
"grad_norm": 2.475069745994417,
|
| 805 |
+
"learning_rate": 8.942807399692543e-06,
|
| 806 |
+
"loss": 0.6244,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2923076923076923,
|
| 811 |
+
"grad_norm": 2.379901357997744,
|
| 812 |
+
"learning_rate": 8.915129312088112e-06,
|
| 813 |
+
"loss": 0.6126,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.2948717948717949,
|
| 818 |
+
"grad_norm": 2.579387941858795,
|
| 819 |
+
"learning_rate": 8.88713758644883e-06,
|
| 820 |
+
"loss": 0.6094,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.29743589743589743,
|
| 825 |
+
"grad_norm": 2.509534211853439,
|
| 826 |
+
"learning_rate": 8.858834465170576e-06,
|
| 827 |
+
"loss": 0.6073,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3,
|
| 832 |
+
"grad_norm": 2.4610011089524164,
|
| 833 |
+
"learning_rate": 8.83022221559489e-06,
|
| 834 |
+
"loss": 0.6018,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.30256410256410254,
|
| 839 |
+
"grad_norm": 2.3560790069702304,
|
| 840 |
+
"learning_rate": 8.801303129827352e-06,
|
| 841 |
+
"loss": 0.5972,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.30512820512820515,
|
| 846 |
+
"grad_norm": 2.564068568417491,
|
| 847 |
+
"learning_rate": 8.772079524553951e-06,
|
| 848 |
+
"loss": 0.6027,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3076923076923077,
|
| 853 |
+
"grad_norm": 2.2846318016556513,
|
| 854 |
+
"learning_rate": 8.742553740855507e-06,
|
| 855 |
+
"loss": 0.5968,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.31025641025641026,
|
| 860 |
+
"grad_norm": 2.3284611207841217,
|
| 861 |
+
"learning_rate": 8.712728144020118e-06,
|
| 862 |
+
"loss": 0.6043,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3128205128205128,
|
| 867 |
+
"grad_norm": 2.3902221577154528,
|
| 868 |
+
"learning_rate": 8.682605123353685e-06,
|
| 869 |
+
"loss": 0.5961,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3153846153846154,
|
| 874 |
+
"grad_norm": 2.425567977054307,
|
| 875 |
+
"learning_rate": 8.652187091988516e-06,
|
| 876 |
+
"loss": 0.5936,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.31794871794871793,
|
| 881 |
+
"grad_norm": 2.4720501112801068,
|
| 882 |
+
"learning_rate": 8.621476486689991e-06,
|
| 883 |
+
"loss": 0.5914,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.32051282051282054,
|
| 888 |
+
"grad_norm": 2.5269541650197063,
|
| 889 |
+
"learning_rate": 8.590475767661371e-06,
|
| 890 |
+
"loss": 0.5825,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3230769230769231,
|
| 895 |
+
"grad_norm": 2.441311110280109,
|
| 896 |
+
"learning_rate": 8.559187418346703e-06,
|
| 897 |
+
"loss": 0.5847,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.32564102564102565,
|
| 902 |
+
"grad_norm": 2.4920344440939575,
|
| 903 |
+
"learning_rate": 8.527613945231886e-06,
|
| 904 |
+
"loss": 0.5849,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3282051282051282,
|
| 909 |
+
"grad_norm": 2.2770162283475615,
|
| 910 |
+
"learning_rate": 8.495757877643857e-06,
|
| 911 |
+
"loss": 0.5775,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.33076923076923076,
|
| 916 |
+
"grad_norm": 2.45173559412692,
|
| 917 |
+
"learning_rate": 8.463621767547998e-06,
|
| 918 |
+
"loss": 0.5714,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3333333333333333,
|
| 923 |
+
"grad_norm": 2.37188665141408,
|
| 924 |
+
"learning_rate": 8.43120818934367e-06,
|
| 925 |
+
"loss": 0.582,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.33589743589743587,
|
| 930 |
+
"grad_norm": 2.28633165733625,
|
| 931 |
+
"learning_rate": 8.398519739657997e-06,
|
| 932 |
+
"loss": 0.5766,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3384615384615385,
|
| 937 |
+
"grad_norm": 2.390940218026475,
|
| 938 |
+
"learning_rate": 8.36555903713785e-06,
|
| 939 |
+
"loss": 0.5824,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.34102564102564104,
|
| 944 |
+
"grad_norm": 2.436136269822359,
|
| 945 |
+
"learning_rate": 8.332328722240072e-06,
|
| 946 |
+
"loss": 0.5801,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3435897435897436,
|
| 951 |
+
"grad_norm": 2.3346149193120183,
|
| 952 |
+
"learning_rate": 8.298831457019943e-06,
|
| 953 |
+
"loss": 0.5781,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.34615384615384615,
|
| 958 |
+
"grad_norm": 2.343996317953977,
|
| 959 |
+
"learning_rate": 8.265069924917925e-06,
|
| 960 |
+
"loss": 0.573,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3487179487179487,
|
| 965 |
+
"grad_norm": 2.362387018042547,
|
| 966 |
+
"learning_rate": 8.231046830544716e-06,
|
| 967 |
+
"loss": 0.5459,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.35128205128205126,
|
| 972 |
+
"grad_norm": 2.3640353741297178,
|
| 973 |
+
"learning_rate": 8.196764899464552e-06,
|
| 974 |
+
"loss": 0.577,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.35384615384615387,
|
| 979 |
+
"grad_norm": 2.393950482821057,
|
| 980 |
+
"learning_rate": 8.162226877976886e-06,
|
| 981 |
+
"loss": 0.571,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.3564102564102564,
|
| 986 |
+
"grad_norm": 2.5266665484244717,
|
| 987 |
+
"learning_rate": 8.127435532896388e-06,
|
| 988 |
+
"loss": 0.5698,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.358974358974359,
|
| 993 |
+
"grad_norm": 2.4236369019301987,
|
| 994 |
+
"learning_rate": 8.092393651331275e-06,
|
| 995 |
+
"loss": 0.5574,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.36153846153846153,
|
| 1000 |
+
"grad_norm": 2.366524077967537,
|
| 1001 |
+
"learning_rate": 8.057104040460062e-06,
|
| 1002 |
+
"loss": 0.5656,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3641025641025641,
|
| 1007 |
+
"grad_norm": 2.51901241992306,
|
| 1008 |
+
"learning_rate": 8.021569527306663e-06,
|
| 1009 |
+
"loss": 0.5564,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.36666666666666664,
|
| 1014 |
+
"grad_norm": 2.3012788394311388,
|
| 1015 |
+
"learning_rate": 7.985792958513932e-06,
|
| 1016 |
+
"loss": 0.5452,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.36923076923076925,
|
| 1021 |
+
"grad_norm": 2.256394871040715,
|
| 1022 |
+
"learning_rate": 7.949777200115617e-06,
|
| 1023 |
+
"loss": 0.5573,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3717948717948718,
|
| 1028 |
+
"grad_norm": 2.345080301865664,
|
| 1029 |
+
"learning_rate": 7.913525137306756e-06,
|
| 1030 |
+
"loss": 0.562,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.37435897435897436,
|
| 1035 |
+
"grad_norm": 2.356034194528926,
|
| 1036 |
+
"learning_rate": 7.877039674212569e-06,
|
| 1037 |
+
"loss": 0.5502,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.3769230769230769,
|
| 1042 |
+
"grad_norm": 2.258668319004814,
|
| 1043 |
+
"learning_rate": 7.84032373365578e-06,
|
| 1044 |
+
"loss": 0.5583,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.37948717948717947,
|
| 1049 |
+
"grad_norm": 2.406002737003363,
|
| 1050 |
+
"learning_rate": 7.803380256922495e-06,
|
| 1051 |
+
"loss": 0.5532,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.382051282051282,
|
| 1056 |
+
"grad_norm": 2.4436486647408557,
|
| 1057 |
+
"learning_rate": 7.76621220352657e-06,
|
| 1058 |
+
"loss": 0.5625,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.38461538461538464,
|
| 1063 |
+
"grad_norm": 2.2622529339051116,
|
| 1064 |
+
"learning_rate": 7.728822550972523e-06,
|
| 1065 |
+
"loss": 0.544,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.3871794871794872,
|
| 1070 |
+
"grad_norm": 2.360201684948899,
|
| 1071 |
+
"learning_rate": 7.69121429451702e-06,
|
| 1072 |
+
"loss": 0.5419,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.38974358974358975,
|
| 1077 |
+
"grad_norm": 2.523997978763824,
|
| 1078 |
+
"learning_rate": 7.65339044692891e-06,
|
| 1079 |
+
"loss": 0.5447,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.3923076923076923,
|
| 1084 |
+
"grad_norm": 2.3781942951542905,
|
| 1085 |
+
"learning_rate": 7.615354038247889e-06,
|
| 1086 |
+
"loss": 0.5455,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.39487179487179486,
|
| 1091 |
+
"grad_norm": 2.3551793216491252,
|
| 1092 |
+
"learning_rate": 7.577108115541761e-06,
|
| 1093 |
+
"loss": 0.5522,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.3974358974358974,
|
| 1098 |
+
"grad_norm": 2.2869590132059687,
|
| 1099 |
+
"learning_rate": 7.53865574266234e-06,
|
| 1100 |
+
"loss": 0.5425,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4,
|
| 1105 |
+
"grad_norm": 2.3895152812865135,
|
| 1106 |
+
"learning_rate": 7.500000000000001e-06,
|
| 1107 |
+
"loss": 0.5546,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4025641025641026,
|
| 1112 |
+
"grad_norm": 2.326678141379568,
|
| 1113 |
+
"learning_rate": 7.461143984236925e-06,
|
| 1114 |
+
"loss": 0.5336,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.40512820512820513,
|
| 1119 |
+
"grad_norm": 2.332360549526255,
|
| 1120 |
+
"learning_rate": 7.422090808099014e-06,
|
| 1121 |
+
"loss": 0.5582,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4076923076923077,
|
| 1126 |
+
"grad_norm": 2.73441064133198,
|
| 1127 |
+
"learning_rate": 7.382843600106539e-06,
|
| 1128 |
+
"loss": 0.5446,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.41025641025641024,
|
| 1133 |
+
"grad_norm": 2.4297671403954126,
|
| 1134 |
+
"learning_rate": 7.343405504323519e-06,
|
| 1135 |
+
"loss": 0.5352,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4128205128205128,
|
| 1140 |
+
"grad_norm": 2.407219494603257,
|
| 1141 |
+
"learning_rate": 7.303779680105844e-06,
|
| 1142 |
+
"loss": 0.5369,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4153846153846154,
|
| 1147 |
+
"grad_norm": 2.409663247487236,
|
| 1148 |
+
"learning_rate": 7.263969301848188e-06,
|
| 1149 |
+
"loss": 0.5235,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.41794871794871796,
|
| 1154 |
+
"grad_norm": 2.4425251774533114,
|
| 1155 |
+
"learning_rate": 7.223977558729707e-06,
|
| 1156 |
+
"loss": 0.5465,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4205128205128205,
|
| 1161 |
+
"grad_norm": 2.288578248006796,
|
| 1162 |
+
"learning_rate": 7.183807654458565e-06,
|
| 1163 |
+
"loss": 0.5143,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4230769230769231,
|
| 1168 |
+
"grad_norm": 2.3832843352346846,
|
| 1169 |
+
"learning_rate": 7.143462807015271e-06,
|
| 1170 |
+
"loss": 0.531,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4256410256410256,
|
| 1175 |
+
"grad_norm": 2.353471391032929,
|
| 1176 |
+
"learning_rate": 7.102946248394908e-06,
|
| 1177 |
+
"loss": 0.5163,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4282051282051282,
|
| 1182 |
+
"grad_norm": 2.328206416879441,
|
| 1183 |
+
"learning_rate": 7.0622612243482035e-06,
|
| 1184 |
+
"loss": 0.5263,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.4307692307692308,
|
| 1189 |
+
"grad_norm": 2.3182487758013672,
|
| 1190 |
+
"learning_rate": 7.021410994121525e-06,
|
| 1191 |
+
"loss": 0.5244,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.43333333333333335,
|
| 1196 |
+
"grad_norm": 2.4570735364008627,
|
| 1197 |
+
"learning_rate": 6.980398830195785e-06,
|
| 1198 |
+
"loss": 0.5268,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.4358974358974359,
|
| 1203 |
+
"grad_norm": 2.363628640743217,
|
| 1204 |
+
"learning_rate": 6.939228018024275e-06,
|
| 1205 |
+
"loss": 0.5327,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.43846153846153846,
|
| 1210 |
+
"grad_norm": 2.532584427272012,
|
| 1211 |
+
"learning_rate": 6.897901855769483e-06,
|
| 1212 |
+
"loss": 0.5043,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.441025641025641,
|
| 1217 |
+
"grad_norm": 2.32923251434474,
|
| 1218 |
+
"learning_rate": 6.856423654038868e-06,
|
| 1219 |
+
"loss": 0.5274,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.44358974358974357,
|
| 1224 |
+
"grad_norm": 2.280683467658302,
|
| 1225 |
+
"learning_rate": 6.814796735619664e-06,
|
| 1226 |
+
"loss": 0.5215,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4461538461538462,
|
| 1231 |
+
"grad_norm": 2.41025108576172,
|
| 1232 |
+
"learning_rate": 6.773024435212678e-06,
|
| 1233 |
+
"loss": 0.5166,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.44871794871794873,
|
| 1238 |
+
"grad_norm": 2.315707168471926,
|
| 1239 |
+
"learning_rate": 6.731110099165165e-06,
|
| 1240 |
+
"loss": 0.515,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4512820512820513,
|
| 1245 |
+
"grad_norm": 276.06259122484,
|
| 1246 |
+
"learning_rate": 6.689057085202737e-06,
|
| 1247 |
+
"loss": 0.5416,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.45384615384615384,
|
| 1252 |
+
"grad_norm": 2.4156201482273016,
|
| 1253 |
+
"learning_rate": 6.646868762160399e-06,
|
| 1254 |
+
"loss": 0.5159,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4564102564102564,
|
| 1259 |
+
"grad_norm": 2.2786600416058924,
|
| 1260 |
+
"learning_rate": 6.6045485097126585e-06,
|
| 1261 |
+
"loss": 0.5101,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.45897435897435895,
|
| 1266 |
+
"grad_norm": 2.353965318631809,
|
| 1267 |
+
"learning_rate": 6.562099718102788e-06,
|
| 1268 |
+
"loss": 0.5155,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.46153846153846156,
|
| 1273 |
+
"grad_norm": 2.4114551311265204,
|
| 1274 |
+
"learning_rate": 6.519525787871235e-06,
|
| 1275 |
+
"loss": 0.517,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4641025641025641,
|
| 1280 |
+
"grad_norm": 2.3985660530296062,
|
| 1281 |
+
"learning_rate": 6.476830129583207e-06,
|
| 1282 |
+
"loss": 0.5004,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4666666666666667,
|
| 1287 |
+
"grad_norm": 2.360463974629649,
|
| 1288 |
+
"learning_rate": 6.434016163555452e-06,
|
| 1289 |
+
"loss": 0.5096,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.46923076923076923,
|
| 1294 |
+
"grad_norm": 2.4536941202621745,
|
| 1295 |
+
"learning_rate": 6.391087319582264e-06,
|
| 1296 |
+
"loss": 0.5134,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4717948717948718,
|
| 1301 |
+
"grad_norm": 2.2791112676567744,
|
| 1302 |
+
"learning_rate": 6.34804703666072e-06,
|
| 1303 |
+
"loss": 0.5146,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.47435897435897434,
|
| 1308 |
+
"grad_norm": 2.3821675561579236,
|
| 1309 |
+
"learning_rate": 6.304898762715187e-06,
|
| 1310 |
+
"loss": 0.508,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.47692307692307695,
|
| 1315 |
+
"grad_norm": 2.4529360596394265,
|
| 1316 |
+
"learning_rate": 6.261645954321109e-06,
|
| 1317 |
+
"loss": 0.4985,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.4794871794871795,
|
| 1322 |
+
"grad_norm": 2.256298131213456,
|
| 1323 |
+
"learning_rate": 6.21829207642811e-06,
|
| 1324 |
+
"loss": 0.4921,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.48205128205128206,
|
| 1329 |
+
"grad_norm": 2.448767917095956,
|
| 1330 |
+
"learning_rate": 6.1748406020824115e-06,
|
| 1331 |
+
"loss": 0.4978,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.4846153846153846,
|
| 1336 |
+
"grad_norm": 2.315987919198238,
|
| 1337 |
+
"learning_rate": 6.131295012148613e-06,
|
| 1338 |
+
"loss": 0.5003,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.48717948717948717,
|
| 1343 |
+
"grad_norm": 2.4549261729386482,
|
| 1344 |
+
"learning_rate": 6.087658795030838e-06,
|
| 1345 |
+
"loss": 0.5119,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.4897435897435897,
|
| 1350 |
+
"grad_norm": 2.3677954185891728,
|
| 1351 |
+
"learning_rate": 6.043935446393294e-06,
|
| 1352 |
+
"loss": 0.4872,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.49230769230769234,
|
| 1357 |
+
"grad_norm": 2.264324311811878,
|
| 1358 |
+
"learning_rate": 6.000128468880223e-06,
|
| 1359 |
+
"loss": 0.5122,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.4948717948717949,
|
| 1364 |
+
"grad_norm": 2.3791395718772006,
|
| 1365 |
+
"learning_rate": 5.956241371835312e-06,
|
| 1366 |
+
"loss": 0.4903,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.49743589743589745,
|
| 1371 |
+
"grad_norm": 2.3183613178906843,
|
| 1372 |
+
"learning_rate": 5.912277671020564e-06,
|
| 1373 |
+
"loss": 0.4977,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5,
|
| 1378 |
+
"grad_norm": 2.29335823998158,
|
| 1379 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 1380 |
+
"loss": 0.4869,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5025641025641026,
|
| 1385 |
+
"grad_norm": 2.3281002741187646,
|
| 1386 |
+
"learning_rate": 5.824134551530783e-06,
|
| 1387 |
+
"loss": 0.4876,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5051282051282051,
|
| 1392 |
+
"grad_norm": 2.5742883235915617,
|
| 1393 |
+
"learning_rate": 5.77996219393409e-06,
|
| 1394 |
+
"loss": 0.4915,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5076923076923077,
|
| 1399 |
+
"grad_norm": 2.365622183252692,
|
| 1400 |
+
"learning_rate": 5.735727354158581e-06,
|
| 1401 |
+
"loss": 0.4805,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5102564102564102,
|
| 1406 |
+
"grad_norm": 2.3390404631244186,
|
| 1407 |
+
"learning_rate": 5.6914335758236665e-06,
|
| 1408 |
+
"loss": 0.4847,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5128205128205128,
|
| 1413 |
+
"grad_norm": 2.409479555334482,
|
| 1414 |
+
"learning_rate": 5.647084407270277e-06,
|
| 1415 |
+
"loss": 0.4803,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5153846153846153,
|
| 1420 |
+
"grad_norm": 2.1772689630729722,
|
| 1421 |
+
"learning_rate": 5.6026834012766155e-06,
|
| 1422 |
+
"loss": 0.4639,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.517948717948718,
|
| 1427 |
+
"grad_norm": 2.344704474447901,
|
| 1428 |
+
"learning_rate": 5.5582341147735396e-06,
|
| 1429 |
+
"loss": 0.4908,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5205128205128206,
|
| 1434 |
+
"grad_norm": 2.377431610687667,
|
| 1435 |
+
"learning_rate": 5.5137401085596224e-06,
|
| 1436 |
+
"loss": 0.4726,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5230769230769231,
|
| 1441 |
+
"grad_norm": 2.232621441545563,
|
| 1442 |
+
"learning_rate": 5.469204947015897e-06,
|
| 1443 |
+
"loss": 0.4781,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5256410256410257,
|
| 1448 |
+
"grad_norm": 2.2830639259821326,
|
| 1449 |
+
"learning_rate": 5.424632197820325e-06,
|
| 1450 |
+
"loss": 0.481,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5282051282051282,
|
| 1455 |
+
"grad_norm": 2.278086265992867,
|
| 1456 |
+
"learning_rate": 5.380025431661981e-06,
|
| 1457 |
+
"loss": 0.4638,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5307692307692308,
|
| 1462 |
+
"grad_norm": 2.269191869325939,
|
| 1463 |
+
"learning_rate": 5.335388221955012e-06,
|
| 1464 |
+
"loss": 0.4832,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5333333333333333,
|
| 1469 |
+
"grad_norm": 2.454836078007937,
|
| 1470 |
+
"learning_rate": 5.290724144552379e-06,
|
| 1471 |
+
"loss": 0.4724,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5358974358974359,
|
| 1476 |
+
"grad_norm": 2.3451732210243286,
|
| 1477 |
+
"learning_rate": 5.246036777459391e-06,
|
| 1478 |
+
"loss": 0.4728,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5384615384615384,
|
| 1483 |
+
"grad_norm": 2.3588173670427595,
|
| 1484 |
+
"learning_rate": 5.201329700547077e-06,
|
| 1485 |
+
"loss": 0.4696,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.541025641025641,
|
| 1490 |
+
"grad_norm": 2.3165391366093684,
|
| 1491 |
+
"learning_rate": 5.156606495265402e-06,
|
| 1492 |
+
"loss": 0.4778,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5435897435897435,
|
| 1497 |
+
"grad_norm": 2.421604099533384,
|
| 1498 |
+
"learning_rate": 5.111870744356366e-06,
|
| 1499 |
+
"loss": 0.4802,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5461538461538461,
|
| 1504 |
+
"grad_norm": 2.437578551376967,
|
| 1505 |
+
"learning_rate": 5.067126031566988e-06,
|
| 1506 |
+
"loss": 0.471,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5487179487179488,
|
| 1511 |
+
"grad_norm": 2.2866015786243197,
|
| 1512 |
+
"learning_rate": 5.022375941362218e-06,
|
| 1513 |
+
"loss": 0.4634,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5512820512820513,
|
| 1518 |
+
"grad_norm": 2.3760034162224057,
|
| 1519 |
+
"learning_rate": 4.977624058637783e-06,
|
| 1520 |
+
"loss": 0.461,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5538461538461539,
|
| 1525 |
+
"grad_norm": 2.196491886505775,
|
| 1526 |
+
"learning_rate": 4.932873968433014e-06,
|
| 1527 |
+
"loss": 0.4613,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5564102564102564,
|
| 1532 |
+
"grad_norm": 2.3547607515039726,
|
| 1533 |
+
"learning_rate": 4.8881292556436355e-06,
|
| 1534 |
+
"loss": 0.4681,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.558974358974359,
|
| 1539 |
+
"grad_norm": 2.3142668747054422,
|
| 1540 |
+
"learning_rate": 4.8433935047346e-06,
|
| 1541 |
+
"loss": 0.4646,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5615384615384615,
|
| 1546 |
+
"grad_norm": 2.2294897374308724,
|
| 1547 |
+
"learning_rate": 4.798670299452926e-06,
|
| 1548 |
+
"loss": 0.4621,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5641025641025641,
|
| 1553 |
+
"grad_norm": 2.201796451540025,
|
| 1554 |
+
"learning_rate": 4.75396322254061e-06,
|
| 1555 |
+
"loss": 0.4643,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5666666666666667,
|
| 1560 |
+
"grad_norm": 2.3004102141787706,
|
| 1561 |
+
"learning_rate": 4.7092758554476215e-06,
|
| 1562 |
+
"loss": 0.4473,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5692307692307692,
|
| 1567 |
+
"grad_norm": 2.4629293661391642,
|
| 1568 |
+
"learning_rate": 4.664611778044988e-06,
|
| 1569 |
+
"loss": 0.4595,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5717948717948718,
|
| 1574 |
+
"grad_norm": 2.3152580137062198,
|
| 1575 |
+
"learning_rate": 4.619974568338021e-06,
|
| 1576 |
+
"loss": 0.4532,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5743589743589743,
|
| 1581 |
+
"grad_norm": 2.30027584904841,
|
| 1582 |
+
"learning_rate": 4.575367802179675e-06,
|
| 1583 |
+
"loss": 0.4455,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5769230769230769,
|
| 1588 |
+
"grad_norm": 2.183997017733211,
|
| 1589 |
+
"learning_rate": 4.530795052984104e-06,
|
| 1590 |
+
"loss": 0.4535,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.5794871794871795,
|
| 1595 |
+
"grad_norm": 2.233588195344829,
|
| 1596 |
+
"learning_rate": 4.48625989144038e-06,
|
| 1597 |
+
"loss": 0.4568,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5820512820512821,
|
| 1602 |
+
"grad_norm": 2.3225007847891375,
|
| 1603 |
+
"learning_rate": 4.441765885226462e-06,
|
| 1604 |
+
"loss": 0.4454,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5846153846153846,
|
| 1609 |
+
"grad_norm": 2.1432323382255887,
|
| 1610 |
+
"learning_rate": 4.397316598723385e-06,
|
| 1611 |
+
"loss": 0.452,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5871794871794872,
|
| 1616 |
+
"grad_norm": 2.249098192197577,
|
| 1617 |
+
"learning_rate": 4.352915592729723e-06,
|
| 1618 |
+
"loss": 0.4406,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5897435897435898,
|
| 1623 |
+
"grad_norm": 2.3668720077166663,
|
| 1624 |
+
"learning_rate": 4.308566424176336e-06,
|
| 1625 |
+
"loss": 0.4612,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5923076923076923,
|
| 1630 |
+
"grad_norm": 2.3280371739332106,
|
| 1631 |
+
"learning_rate": 4.264272645841419e-06,
|
| 1632 |
+
"loss": 0.4563,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.5948717948717949,
|
| 1637 |
+
"grad_norm": 2.2992039199150343,
|
| 1638 |
+
"learning_rate": 4.220037806065911e-06,
|
| 1639 |
+
"loss": 0.4404,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.5974358974358974,
|
| 1644 |
+
"grad_norm": 2.352728324858546,
|
| 1645 |
+
"learning_rate": 4.175865448469219e-06,
|
| 1646 |
+
"loss": 0.447,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6,
|
| 1651 |
+
"grad_norm": 2.2680523365737466,
|
| 1652 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1653 |
+
"loss": 0.4374,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6025641025641025,
|
| 1658 |
+
"grad_norm": 2.339304021822968,
|
| 1659 |
+
"learning_rate": 4.087722328979437e-06,
|
| 1660 |
+
"loss": 0.4576,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6051282051282051,
|
| 1665 |
+
"grad_norm": 2.360150826251379,
|
| 1666 |
+
"learning_rate": 4.043758628164688e-06,
|
| 1667 |
+
"loss": 0.437,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6076923076923076,
|
| 1672 |
+
"grad_norm": 2.3068618660691937,
|
| 1673 |
+
"learning_rate": 3.999871531119779e-06,
|
| 1674 |
+
"loss": 0.4336,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6102564102564103,
|
| 1679 |
+
"grad_norm": 2.2504790874563785,
|
| 1680 |
+
"learning_rate": 3.956064553606708e-06,
|
| 1681 |
+
"loss": 0.4303,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6128205128205129,
|
| 1686 |
+
"grad_norm": 2.3568871454086784,
|
| 1687 |
+
"learning_rate": 3.912341204969164e-06,
|
| 1688 |
+
"loss": 0.4505,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6153846153846154,
|
| 1693 |
+
"grad_norm": 2.29775545667767,
|
| 1694 |
+
"learning_rate": 3.86870498785139e-06,
|
| 1695 |
+
"loss": 0.4349,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.617948717948718,
|
| 1700 |
+
"grad_norm": 2.249325620803784,
|
| 1701 |
+
"learning_rate": 3.825159397917589e-06,
|
| 1702 |
+
"loss": 0.4245,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6205128205128205,
|
| 1707 |
+
"grad_norm": 2.217056799875781,
|
| 1708 |
+
"learning_rate": 3.781707923571891e-06,
|
| 1709 |
+
"loss": 0.4409,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6230769230769231,
|
| 1714 |
+
"grad_norm": 2.162077945290354,
|
| 1715 |
+
"learning_rate": 3.7383540456788915e-06,
|
| 1716 |
+
"loss": 0.4321,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6256410256410256,
|
| 1721 |
+
"grad_norm": 2.269939213889014,
|
| 1722 |
+
"learning_rate": 3.695101237284815e-06,
|
| 1723 |
+
"loss": 0.4323,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6282051282051282,
|
| 1728 |
+
"grad_norm": 2.321141003845967,
|
| 1729 |
+
"learning_rate": 3.6519529633392825e-06,
|
| 1730 |
+
"loss": 0.4307,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6307692307692307,
|
| 1735 |
+
"grad_norm": 2.290163674183007,
|
| 1736 |
+
"learning_rate": 3.6089126804177373e-06,
|
| 1737 |
+
"loss": 0.43,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6333333333333333,
|
| 1742 |
+
"grad_norm": 2.2796933735948235,
|
| 1743 |
+
"learning_rate": 3.5659838364445505e-06,
|
| 1744 |
+
"loss": 0.4257,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6358974358974359,
|
| 1749 |
+
"grad_norm": 2.3515817559831516,
|
| 1750 |
+
"learning_rate": 3.523169870416795e-06,
|
| 1751 |
+
"loss": 0.421,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6384615384615384,
|
| 1756 |
+
"grad_norm": 2.224007354395459,
|
| 1757 |
+
"learning_rate": 3.480474212128766e-06,
|
| 1758 |
+
"loss": 0.4371,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6410256410256411,
|
| 1763 |
+
"grad_norm": 2.2228662307484472,
|
| 1764 |
+
"learning_rate": 3.4379002818972122e-06,
|
| 1765 |
+
"loss": 0.4251,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6435897435897436,
|
| 1770 |
+
"grad_norm": 2.354244944899084,
|
| 1771 |
+
"learning_rate": 3.3954514902873427e-06,
|
| 1772 |
+
"loss": 0.4329,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6461538461538462,
|
| 1777 |
+
"grad_norm": 2.2750993825918093,
|
| 1778 |
+
"learning_rate": 3.3531312378396026e-06,
|
| 1779 |
+
"loss": 0.4151,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6487179487179487,
|
| 1784 |
+
"grad_norm": 2.2325391828770274,
|
| 1785 |
+
"learning_rate": 3.310942914797265e-06,
|
| 1786 |
+
"loss": 0.4327,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6512820512820513,
|
| 1791 |
+
"grad_norm": 2.3345273581362935,
|
| 1792 |
+
"learning_rate": 3.2688899008348386e-06,
|
| 1793 |
+
"loss": 0.4111,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6538461538461539,
|
| 1798 |
+
"grad_norm": 2.3549983976092284,
|
| 1799 |
+
"learning_rate": 3.226975564787322e-06,
|
| 1800 |
+
"loss": 0.4182,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6564102564102564,
|
| 1805 |
+
"grad_norm": 2.2210755036476924,
|
| 1806 |
+
"learning_rate": 3.1852032643803377e-06,
|
| 1807 |
+
"loss": 0.4292,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.658974358974359,
|
| 1812 |
+
"grad_norm": 2.3446968695245376,
|
| 1813 |
+
"learning_rate": 3.143576345961132e-06,
|
| 1814 |
+
"loss": 0.4173,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6615384615384615,
|
| 1819 |
+
"grad_norm": 2.2100916363822765,
|
| 1820 |
+
"learning_rate": 3.1020981442305187e-06,
|
| 1821 |
+
"loss": 0.4103,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6641025641025641,
|
| 1826 |
+
"grad_norm": 2.3879357481967207,
|
| 1827 |
+
"learning_rate": 3.0607719819757264e-06,
|
| 1828 |
+
"loss": 0.4249,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6666666666666666,
|
| 1833 |
+
"grad_norm": 2.321287867020548,
|
| 1834 |
+
"learning_rate": 3.019601169804216e-06,
|
| 1835 |
+
"loss": 0.4291,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6692307692307692,
|
| 1840 |
+
"grad_norm": 2.1709645290877475,
|
| 1841 |
+
"learning_rate": 2.978589005878476e-06,
|
| 1842 |
+
"loss": 0.4043,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6717948717948717,
|
| 1847 |
+
"grad_norm": 2.1963788275980867,
|
| 1848 |
+
"learning_rate": 2.937738775651798e-06,
|
| 1849 |
+
"loss": 0.4195,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6743589743589744,
|
| 1854 |
+
"grad_norm": 2.3134447029337277,
|
| 1855 |
+
"learning_rate": 2.8970537516050935e-06,
|
| 1856 |
+
"loss": 0.415,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.676923076923077,
|
| 1861 |
+
"grad_norm": 2.215720987796974,
|
| 1862 |
+
"learning_rate": 2.8565371929847286e-06,
|
| 1863 |
+
"loss": 0.4168,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6794871794871795,
|
| 1868 |
+
"grad_norm": 2.4752032392325716,
|
| 1869 |
+
"learning_rate": 2.816192345541437e-06,
|
| 1870 |
+
"loss": 0.4181,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6820512820512821,
|
| 1875 |
+
"grad_norm": 2.201152450487114,
|
| 1876 |
+
"learning_rate": 2.776022441270295e-06,
|
| 1877 |
+
"loss": 0.4156,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6846153846153846,
|
| 1882 |
+
"grad_norm": 2.2072519693672272,
|
| 1883 |
+
"learning_rate": 2.736030698151815e-06,
|
| 1884 |
+
"loss": 0.4166,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6871794871794872,
|
| 1889 |
+
"grad_norm": 2.3733944621945464,
|
| 1890 |
+
"learning_rate": 2.6962203198941587e-06,
|
| 1891 |
+
"loss": 0.4191,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6897435897435897,
|
| 1896 |
+
"grad_norm": 2.3531096957130746,
|
| 1897 |
+
"learning_rate": 2.656594495676482e-06,
|
| 1898 |
+
"loss": 0.4082,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6923076923076923,
|
| 1903 |
+
"grad_norm": 2.180269591424075,
|
| 1904 |
+
"learning_rate": 2.6171563998934605e-06,
|
| 1905 |
+
"loss": 0.3952,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6948717948717948,
|
| 1910 |
+
"grad_norm": 2.066368640585333,
|
| 1911 |
+
"learning_rate": 2.577909191900988e-06,
|
| 1912 |
+
"loss": 0.4192,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6974358974358974,
|
| 1917 |
+
"grad_norm": 2.2737966608576534,
|
| 1918 |
+
"learning_rate": 2.5388560157630765e-06,
|
| 1919 |
+
"loss": 0.3935,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7,
|
| 1924 |
+
"grad_norm": 2.17705872571539,
|
| 1925 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1926 |
+
"loss": 0.4125,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7025641025641025,
|
| 1931 |
+
"grad_norm": 2.218745613533214,
|
| 1932 |
+
"learning_rate": 2.4613442573376625e-06,
|
| 1933 |
+
"loss": 0.4008,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7051282051282052,
|
| 1938 |
+
"grad_norm": 2.1239348156972016,
|
| 1939 |
+
"learning_rate": 2.422891884458241e-06,
|
| 1940 |
+
"loss": 0.4094,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7076923076923077,
|
| 1945 |
+
"grad_norm": 2.179223646983379,
|
| 1946 |
+
"learning_rate": 2.384645961752113e-06,
|
| 1947 |
+
"loss": 0.4092,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7102564102564103,
|
| 1952 |
+
"grad_norm": 2.3185213442449406,
|
| 1953 |
+
"learning_rate": 2.346609553071093e-06,
|
| 1954 |
+
"loss": 0.4088,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7128205128205128,
|
| 1959 |
+
"grad_norm": 2.294944820481849,
|
| 1960 |
+
"learning_rate": 2.308785705482982e-06,
|
| 1961 |
+
"loss": 0.3894,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7153846153846154,
|
| 1966 |
+
"grad_norm": 2.1876725669932,
|
| 1967 |
+
"learning_rate": 2.2711774490274767e-06,
|
| 1968 |
+
"loss": 0.3913,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.717948717948718,
|
| 1973 |
+
"grad_norm": 2.3670833638473607,
|
| 1974 |
+
"learning_rate": 2.2337877964734324e-06,
|
| 1975 |
+
"loss": 0.4004,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7205128205128205,
|
| 1980 |
+
"grad_norm": 2.213035848099544,
|
| 1981 |
+
"learning_rate": 2.1966197430775056e-06,
|
| 1982 |
+
"loss": 0.4051,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7230769230769231,
|
| 1987 |
+
"grad_norm": 2.1456079138282216,
|
| 1988 |
+
"learning_rate": 2.159676266344222e-06,
|
| 1989 |
+
"loss": 0.3998,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7256410256410256,
|
| 1994 |
+
"grad_norm": 2.368556775998427,
|
| 1995 |
+
"learning_rate": 2.122960325787432e-06,
|
| 1996 |
+
"loss": 0.3999,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7282051282051282,
|
| 2001 |
+
"grad_norm": 2.2866531656271714,
|
| 2002 |
+
"learning_rate": 2.086474862693244e-06,
|
| 2003 |
+
"loss": 0.3954,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7307692307692307,
|
| 2008 |
+
"grad_norm": 2.2147361106092247,
|
| 2009 |
+
"learning_rate": 2.050222799884387e-06,
|
| 2010 |
+
"loss": 0.3973,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7333333333333333,
|
| 2015 |
+
"grad_norm": 2.262377192608549,
|
| 2016 |
+
"learning_rate": 2.0142070414860704e-06,
|
| 2017 |
+
"loss": 0.3858,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.735897435897436,
|
| 2022 |
+
"grad_norm": 2.254959040727779,
|
| 2023 |
+
"learning_rate": 1.9784304726933384e-06,
|
| 2024 |
+
"loss": 0.3957,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7384615384615385,
|
| 2029 |
+
"grad_norm": 2.443844940991985,
|
| 2030 |
+
"learning_rate": 1.942895959539939e-06,
|
| 2031 |
+
"loss": 0.4015,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7410256410256411,
|
| 2036 |
+
"grad_norm": 2.1875651185476577,
|
| 2037 |
+
"learning_rate": 1.9076063486687256e-06,
|
| 2038 |
+
"loss": 0.3822,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7435897435897436,
|
| 2043 |
+
"grad_norm": 2.2266048608412574,
|
| 2044 |
+
"learning_rate": 1.8725644671036125e-06,
|
| 2045 |
+
"loss": 0.4001,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7461538461538462,
|
| 2050 |
+
"grad_norm": 2.142836559910304,
|
| 2051 |
+
"learning_rate": 1.8377731220231144e-06,
|
| 2052 |
+
"loss": 0.3904,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7487179487179487,
|
| 2057 |
+
"grad_norm": 2.2609455403429557,
|
| 2058 |
+
"learning_rate": 1.803235100535452e-06,
|
| 2059 |
+
"loss": 0.3948,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7512820512820513,
|
| 2064 |
+
"grad_norm": 2.264696810926414,
|
| 2065 |
+
"learning_rate": 1.7689531694552863e-06,
|
| 2066 |
+
"loss": 0.3976,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7538461538461538,
|
| 2071 |
+
"grad_norm": 2.230112130398212,
|
| 2072 |
+
"learning_rate": 1.7349300750820758e-06,
|
| 2073 |
+
"loss": 0.3859,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7564102564102564,
|
| 2078 |
+
"grad_norm": 2.2268561992720173,
|
| 2079 |
+
"learning_rate": 1.7011685429800596e-06,
|
| 2080 |
+
"loss": 0.3849,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7589743589743589,
|
| 2085 |
+
"grad_norm": 2.1244858256821306,
|
| 2086 |
+
"learning_rate": 1.6676712777599275e-06,
|
| 2087 |
+
"loss": 0.3835,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7615384615384615,
|
| 2092 |
+
"grad_norm": 2.3007192861817534,
|
| 2093 |
+
"learning_rate": 1.6344409628621482e-06,
|
| 2094 |
+
"loss": 0.3818,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.764102564102564,
|
| 2099 |
+
"grad_norm": 2.1550992793361896,
|
| 2100 |
+
"learning_rate": 1.6014802603420044e-06,
|
| 2101 |
+
"loss": 0.3788,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7666666666666667,
|
| 2106 |
+
"grad_norm": 2.285182407331258,
|
| 2107 |
+
"learning_rate": 1.5687918106563326e-06,
|
| 2108 |
+
"loss": 0.3707,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7692307692307693,
|
| 2113 |
+
"grad_norm": 2.184234648354467,
|
| 2114 |
+
"learning_rate": 1.5363782324520033e-06,
|
| 2115 |
+
"loss": 0.3804,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7717948717948718,
|
| 2120 |
+
"grad_norm": 2.3833866092139546,
|
| 2121 |
+
"learning_rate": 1.504242122356143e-06,
|
| 2122 |
+
"loss": 0.3917,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7743589743589744,
|
| 2127 |
+
"grad_norm": 2.213841954806635,
|
| 2128 |
+
"learning_rate": 1.4723860547681163e-06,
|
| 2129 |
+
"loss": 0.3948,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7769230769230769,
|
| 2134 |
+
"grad_norm": 2.1343451977199335,
|
| 2135 |
+
"learning_rate": 1.4408125816532981e-06,
|
| 2136 |
+
"loss": 0.383,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7794871794871795,
|
| 2141 |
+
"grad_norm": 2.3514768336870824,
|
| 2142 |
+
"learning_rate": 1.4095242323386305e-06,
|
| 2143 |
+
"loss": 0.3989,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.782051282051282,
|
| 2148 |
+
"grad_norm": 2.316285751451528,
|
| 2149 |
+
"learning_rate": 1.3785235133100088e-06,
|
| 2150 |
+
"loss": 0.3836,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7846153846153846,
|
| 2155 |
+
"grad_norm": 2.136192342183207,
|
| 2156 |
+
"learning_rate": 1.347812908011485e-06,
|
| 2157 |
+
"loss": 0.3737,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7871794871794872,
|
| 2162 |
+
"grad_norm": 2.240858366870099,
|
| 2163 |
+
"learning_rate": 1.3173948766463146e-06,
|
| 2164 |
+
"loss": 0.3832,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7897435897435897,
|
| 2169 |
+
"grad_norm": 2.130447099646043,
|
| 2170 |
+
"learning_rate": 1.2872718559798852e-06,
|
| 2171 |
+
"loss": 0.3809,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7923076923076923,
|
| 2176 |
+
"grad_norm": 2.2629722022590615,
|
| 2177 |
+
"learning_rate": 1.257446259144494e-06,
|
| 2178 |
+
"loss": 0.3827,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.7948717948717948,
|
| 2183 |
+
"grad_norm": 2.1714635036854792,
|
| 2184 |
+
"learning_rate": 1.2279204754460494e-06,
|
| 2185 |
+
"loss": 0.3773,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.7974358974358975,
|
| 2190 |
+
"grad_norm": 2.375860318741745,
|
| 2191 |
+
"learning_rate": 1.1986968701726492e-06,
|
| 2192 |
+
"loss": 0.4044,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8,
|
| 2197 |
+
"grad_norm": 2.160587779645344,
|
| 2198 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 2199 |
+
"loss": 0.3875,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8025641025641026,
|
| 2204 |
+
"grad_norm": 2.198147969238877,
|
| 2205 |
+
"learning_rate": 1.141165534829425e-06,
|
| 2206 |
+
"loss": 0.3981,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8051282051282052,
|
| 2211 |
+
"grad_norm": 2.1389541240218373,
|
| 2212 |
+
"learning_rate": 1.1128624135511712e-06,
|
| 2213 |
+
"loss": 0.3861,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8076923076923077,
|
| 2218 |
+
"grad_norm": 2.0625001013798068,
|
| 2219 |
+
"learning_rate": 1.0848706879118893e-06,
|
| 2220 |
+
"loss": 0.3744,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8102564102564103,
|
| 2225 |
+
"grad_norm": 2.255946436278033,
|
| 2226 |
+
"learning_rate": 1.057192600307456e-06,
|
| 2227 |
+
"loss": 0.3663,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8128205128205128,
|
| 2232 |
+
"grad_norm": 2.057286421099507,
|
| 2233 |
+
"learning_rate": 1.0298303680084448e-06,
|
| 2234 |
+
"loss": 0.3806,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8153846153846154,
|
| 2239 |
+
"grad_norm": 2.1422638338742725,
|
| 2240 |
+
"learning_rate": 1.0027861829824953e-06,
|
| 2241 |
+
"loss": 0.3704,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8179487179487179,
|
| 2246 |
+
"grad_norm": 2.018897710780144,
|
| 2247 |
+
"learning_rate": 9.760622117187234e-07,
|
| 2248 |
+
"loss": 0.3759,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8205128205128205,
|
| 2253 |
+
"grad_norm": 2.248733412879705,
|
| 2254 |
+
"learning_rate": 9.496605950541676e-07,
|
| 2255 |
+
"loss": 0.3736,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.823076923076923,
|
| 2260 |
+
"grad_norm": 2.2165304029183894,
|
| 2261 |
+
"learning_rate": 9.235834480022788e-07,
|
| 2262 |
+
"loss": 0.3664,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8256410256410256,
|
| 2267 |
+
"grad_norm": 2.092232769500133,
|
| 2268 |
+
"learning_rate": 8.978328595834984e-07,
|
| 2269 |
+
"loss": 0.3946,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8282051282051283,
|
| 2274 |
+
"grad_norm": 2.051244434233085,
|
| 2275 |
+
"learning_rate": 8.724108926579e-07,
|
| 2276 |
+
"loss": 0.3711,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8307692307692308,
|
| 2281 |
+
"grad_norm": 2.1357191883411146,
|
| 2282 |
+
"learning_rate": 8.473195837599419e-07,
|
| 2283 |
+
"loss": 0.375,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8333333333333334,
|
| 2288 |
+
"grad_norm": 2.3476361797899727,
|
| 2289 |
+
"learning_rate": 8.225609429353187e-07,
|
| 2290 |
+
"loss": 0.379,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8358974358974359,
|
| 2295 |
+
"grad_norm": 2.1434969249566267,
|
| 2296 |
+
"learning_rate": 7.981369535799354e-07,
|
| 2297 |
+
"loss": 0.3718,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8384615384615385,
|
| 2302 |
+
"grad_norm": 2.1291787139915734,
|
| 2303 |
+
"learning_rate": 7.740495722810271e-07,
|
| 2304 |
+
"loss": 0.3619,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.841025641025641,
|
| 2309 |
+
"grad_norm": 2.1371175381872067,
|
| 2310 |
+
"learning_rate": 7.50300728660407e-07,
|
| 2311 |
+
"loss": 0.3836,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8435897435897436,
|
| 2316 |
+
"grad_norm": 2.0182192104108205,
|
| 2317 |
+
"learning_rate": 7.26892325219899e-07,
|
| 2318 |
+
"loss": 0.3696,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8461538461538461,
|
| 2323 |
+
"grad_norm": 2.182781125364085,
|
| 2324 |
+
"learning_rate": 7.03826237188916e-07,
|
| 2325 |
+
"loss": 0.3798,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8487179487179487,
|
| 2330 |
+
"grad_norm": 2.1819095950658967,
|
| 2331 |
+
"learning_rate": 6.811043123742494e-07,
|
| 2332 |
+
"loss": 0.3626,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8512820512820513,
|
| 2337 |
+
"grad_norm": 1.9947387983712035,
|
| 2338 |
+
"learning_rate": 6.587283710120324e-07,
|
| 2339 |
+
"loss": 0.3654,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8538461538461538,
|
| 2344 |
+
"grad_norm": 2.3863133588522447,
|
| 2345 |
+
"learning_rate": 6.367002056219285e-07,
|
| 2346 |
+
"loss": 0.3769,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8564102564102564,
|
| 2351 |
+
"grad_norm": 2.131117707340598,
|
| 2352 |
+
"learning_rate": 6.150215808635334e-07,
|
| 2353 |
+
"loss": 0.3695,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8589743589743589,
|
| 2358 |
+
"grad_norm": 2.127014385149758,
|
| 2359 |
+
"learning_rate": 5.936942333950063e-07,
|
| 2360 |
+
"loss": 0.3689,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8615384615384616,
|
| 2365 |
+
"grad_norm": 2.048791526125599,
|
| 2366 |
+
"learning_rate": 5.727198717339511e-07,
|
| 2367 |
+
"loss": 0.3557,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8641025641025641,
|
| 2372 |
+
"grad_norm": 2.1611061050109286,
|
| 2373 |
+
"learning_rate": 5.521001761205441e-07,
|
| 2374 |
+
"loss": 0.3744,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8666666666666667,
|
| 2379 |
+
"grad_norm": 2.253709876904671,
|
| 2380 |
+
"learning_rate": 5.318367983829393e-07,
|
| 2381 |
+
"loss": 0.3652,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8692307692307693,
|
| 2386 |
+
"grad_norm": 2.0719950658447823,
|
| 2387 |
+
"learning_rate": 5.119313618049309e-07,
|
| 2388 |
+
"loss": 0.3665,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8717948717948718,
|
| 2393 |
+
"grad_norm": 2.0316305277614775,
|
| 2394 |
+
"learning_rate": 4.9238546099592e-07,
|
| 2395 |
+
"loss": 0.3657,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8743589743589744,
|
| 2400 |
+
"grad_norm": 2.5087707696707326,
|
| 2401 |
+
"learning_rate": 4.732006617631729e-07,
|
| 2402 |
+
"loss": 0.3795,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.8769230769230769,
|
| 2407 |
+
"grad_norm": 2.1498909379980287,
|
| 2408 |
+
"learning_rate": 4.54378500986381e-07,
|
| 2409 |
+
"loss": 0.3663,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.8794871794871795,
|
| 2414 |
+
"grad_norm": 1.9681404281106394,
|
| 2415 |
+
"learning_rate": 4.35920486494546e-07,
|
| 2416 |
+
"loss": 0.3754,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.882051282051282,
|
| 2421 |
+
"grad_norm": 2.1003321632219643,
|
| 2422 |
+
"learning_rate": 4.1782809694518533e-07,
|
| 2423 |
+
"loss": 0.3553,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8846153846153846,
|
| 2428 |
+
"grad_norm": 2.0527201502010235,
|
| 2429 |
+
"learning_rate": 4.001027817058789e-07,
|
| 2430 |
+
"loss": 0.371,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.8871794871794871,
|
| 2435 |
+
"grad_norm": 2.201330552650921,
|
| 2436 |
+
"learning_rate": 3.8274596073816784e-07,
|
| 2437 |
+
"loss": 0.3667,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8897435897435897,
|
| 2442 |
+
"grad_norm": 2.047245403408523,
|
| 2443 |
+
"learning_rate": 3.657590244837911e-07,
|
| 2444 |
+
"loss": 0.3557,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.8923076923076924,
|
| 2449 |
+
"grad_norm": 2.227196360471304,
|
| 2450 |
+
"learning_rate": 3.49143333753309e-07,
|
| 2451 |
+
"loss": 0.3699,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.8948717948717949,
|
| 2456 |
+
"grad_norm": 2.1749366287892804,
|
| 2457 |
+
"learning_rate": 3.3290021961708163e-07,
|
| 2458 |
+
"loss": 0.3583,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.8974358974358975,
|
| 2463 |
+
"grad_norm": 2.2245295170216397,
|
| 2464 |
+
"learning_rate": 3.1703098329864237e-07,
|
| 2465 |
+
"loss": 0.3613,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9,
|
| 2470 |
+
"grad_norm": 2.0951082558511125,
|
| 2471 |
+
"learning_rate": 3.015368960704584e-07,
|
| 2472 |
+
"loss": 0.3502,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9025641025641026,
|
| 2477 |
+
"grad_norm": 2.3736518735276277,
|
| 2478 |
+
"learning_rate": 2.864191991520848e-07,
|
| 2479 |
+
"loss": 0.3648,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9051282051282051,
|
| 2484 |
+
"grad_norm": 1.9061212020960874,
|
| 2485 |
+
"learning_rate": 2.71679103610738e-07,
|
| 2486 |
+
"loss": 0.3616,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9076923076923077,
|
| 2491 |
+
"grad_norm": 2.2330643798467213,
|
| 2492 |
+
"learning_rate": 2.573177902642726e-07,
|
| 2493 |
+
"loss": 0.3612,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9102564102564102,
|
| 2498 |
+
"grad_norm": 2.3165257034110054,
|
| 2499 |
+
"learning_rate": 2.4333640958659144e-07,
|
| 2500 |
+
"loss": 0.3636,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9128205128205128,
|
| 2505 |
+
"grad_norm": 2.0684872184860614,
|
| 2506 |
+
"learning_rate": 2.2973608161547755e-07,
|
| 2507 |
+
"loss": 0.3613,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9153846153846154,
|
| 2512 |
+
"grad_norm": 2.247620069960479,
|
| 2513 |
+
"learning_rate": 2.1651789586287442e-07,
|
| 2514 |
+
"loss": 0.3732,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9179487179487179,
|
| 2519 |
+
"grad_norm": 2.0738095357541653,
|
| 2520 |
+
"learning_rate": 2.0368291122759898e-07,
|
| 2521 |
+
"loss": 0.3544,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9205128205128205,
|
| 2526 |
+
"grad_norm": 2.3019916845092947,
|
| 2527 |
+
"learning_rate": 1.9123215591052014e-07,
|
| 2528 |
+
"loss": 0.3618,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9230769230769231,
|
| 2533 |
+
"grad_norm": 2.038724564899476,
|
| 2534 |
+
"learning_rate": 1.7916662733218848e-07,
|
| 2535 |
+
"loss": 0.3801,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9256410256410257,
|
| 2540 |
+
"grad_norm": 2.2889296400333654,
|
| 2541 |
+
"learning_rate": 1.6748729205293024e-07,
|
| 2542 |
+
"loss": 0.3712,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9282051282051282,
|
| 2547 |
+
"grad_norm": 1.9287399651858075,
|
| 2548 |
+
"learning_rate": 1.5619508569542363e-07,
|
| 2549 |
+
"loss": 0.3616,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9307692307692308,
|
| 2554 |
+
"grad_norm": 1.9456575180320899,
|
| 2555 |
+
"learning_rate": 1.4529091286973994e-07,
|
| 2556 |
+
"loss": 0.369,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9333333333333333,
|
| 2561 |
+
"grad_norm": 2.0665497215433226,
|
| 2562 |
+
"learning_rate": 1.3477564710088097e-07,
|
| 2563 |
+
"loss": 0.3497,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9358974358974359,
|
| 2568 |
+
"grad_norm": 2.0871154036132604,
|
| 2569 |
+
"learning_rate": 1.2465013075879884e-07,
|
| 2570 |
+
"loss": 0.3559,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9384615384615385,
|
| 2575 |
+
"grad_norm": 1.9712698998583276,
|
| 2576 |
+
"learning_rate": 1.1491517499091498e-07,
|
| 2577 |
+
"loss": 0.3541,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.941025641025641,
|
| 2582 |
+
"grad_norm": 2.163801366341123,
|
| 2583 |
+
"learning_rate": 1.055715596571405e-07,
|
| 2584 |
+
"loss": 0.3661,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9435897435897436,
|
| 2589 |
+
"grad_norm": 2.2509599324394984,
|
| 2590 |
+
"learning_rate": 9.662003326740166e-08,
|
| 2591 |
+
"loss": 0.3621,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9461538461538461,
|
| 2596 |
+
"grad_norm": 2.1871133765282345,
|
| 2597 |
+
"learning_rate": 8.80613129216762e-08,
|
| 2598 |
+
"loss": 0.361,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9487179487179487,
|
| 2603 |
+
"grad_norm": 2.1099575248965152,
|
| 2604 |
+
"learning_rate": 7.989608425254924e-08,
|
| 2605 |
+
"loss": 0.3533,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9512820512820512,
|
| 2610 |
+
"grad_norm": 2.176020883382461,
|
| 2611 |
+
"learning_rate": 7.212500137028789e-08,
|
| 2612 |
+
"loss": 0.3678,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9538461538461539,
|
| 2617 |
+
"grad_norm": 2.1204394793153067,
|
| 2618 |
+
"learning_rate": 6.474868681043578e-08,
|
| 2619 |
+
"loss": 0.3526,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9564102564102565,
|
| 2624 |
+
"grad_norm": 2.1621021084689507,
|
| 2625 |
+
"learning_rate": 5.776773148394976e-08,
|
| 2626 |
+
"loss": 0.3651,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.958974358974359,
|
| 2631 |
+
"grad_norm": 2.031947785174627,
|
| 2632 |
+
"learning_rate": 5.1182694629857145e-08,
|
| 2633 |
+
"loss": 0.3734,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9615384615384616,
|
| 2638 |
+
"grad_norm": 2.096096759539924,
|
| 2639 |
+
"learning_rate": 4.499410377045765e-08,
|
| 2640 |
+
"loss": 0.3703,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9641025641025641,
|
| 2645 |
+
"grad_norm": 2.3222972214818283,
|
| 2646 |
+
"learning_rate": 3.9202454669063915e-08,
|
| 2647 |
+
"loss": 0.368,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9666666666666667,
|
| 2652 |
+
"grad_norm": 1.9780233715193638,
|
| 2653 |
+
"learning_rate": 3.3808211290284886e-08,
|
| 2654 |
+
"loss": 0.3479,
|
| 2655 |
+
"step": 1885
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.9692307692307692,
|
| 2659 |
+
"grad_norm": 1.9718006540884354,
|
| 2660 |
+
"learning_rate": 2.8811805762860578e-08,
|
| 2661 |
+
"loss": 0.3555,
|
| 2662 |
+
"step": 1890
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9717948717948718,
|
| 2666 |
+
"grad_norm": 2.2122182201885394,
|
| 2667 |
+
"learning_rate": 2.4213638345040868e-08,
|
| 2668 |
+
"loss": 0.3633,
|
| 2669 |
+
"step": 1895
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.9743589743589743,
|
| 2673 |
+
"grad_norm": 2.0210334835222876,
|
| 2674 |
+
"learning_rate": 2.0014077392525035e-08,
|
| 2675 |
+
"loss": 0.3664,
|
| 2676 |
+
"step": 1900
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9769230769230769,
|
| 2680 |
+
"grad_norm": 1.9777551247351057,
|
| 2681 |
+
"learning_rate": 1.6213459328950355e-08,
|
| 2682 |
+
"loss": 0.3637,
|
| 2683 |
+
"step": 1905
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.9794871794871794,
|
| 2687 |
+
"grad_norm": 1.9849050081166892,
|
| 2688 |
+
"learning_rate": 1.2812088618942009e-08,
|
| 2689 |
+
"loss": 0.3442,
|
| 2690 |
+
"step": 1910
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.982051282051282,
|
| 2694 |
+
"grad_norm": 2.0028405839895775,
|
| 2695 |
+
"learning_rate": 9.810237743724805e-09,
|
| 2696 |
+
"loss": 0.3502,
|
| 2697 |
+
"step": 1915
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.9846153846153847,
|
| 2701 |
+
"grad_norm": 1.982479484802511,
|
| 2702 |
+
"learning_rate": 7.2081471792911914e-09,
|
| 2703 |
+
"loss": 0.3563,
|
| 2704 |
+
"step": 1920
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.9871794871794872,
|
| 2708 |
+
"grad_norm": 2.108119859172211,
|
| 2709 |
+
"learning_rate": 5.006025377138901e-09,
|
| 2710 |
+
"loss": 0.3593,
|
| 2711 |
+
"step": 1925
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.9897435897435898,
|
| 2715 |
+
"grad_norm": 2.1025186070928763,
|
| 2716 |
+
"learning_rate": 3.204048747573185e-09,
|
| 2717 |
+
"loss": 0.36,
|
| 2718 |
+
"step": 1930
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.9923076923076923,
|
| 2722 |
+
"grad_norm": 2.1697555086596823,
|
| 2723 |
+
"learning_rate": 1.8023616455731253e-09,
|
| 2724 |
+
"loss": 0.3606,
|
| 2725 |
+
"step": 1935
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.9948717948717949,
|
| 2729 |
+
"grad_norm": 2.0730870494656424,
|
| 2730 |
+
"learning_rate": 8.010763592264381e-10,
|
| 2731 |
+
"loss": 0.3671,
|
| 2732 |
+
"step": 1940
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.9974358974358974,
|
| 2736 |
+
"grad_norm": 2.131587781280676,
|
| 2737 |
+
"learning_rate": 2.0027310073833516e-10,
|
| 2738 |
+
"loss": 0.3723,
|
| 2739 |
+
"step": 1945
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 1.0,
|
| 2743 |
+
"grad_norm": 1.9834898868136892,
|
| 2744 |
+
"learning_rate": 0.0,
|
| 2745 |
+
"loss": 0.3656,
|
| 2746 |
+
"step": 1950
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 1.0,
|
| 2750 |
+
"eval_loss": 0.3188657760620117,
|
| 2751 |
+
"eval_runtime": 1.193,
|
| 2752 |
+
"eval_samples_per_second": 1.676,
|
| 2753 |
+
"eval_steps_per_second": 0.838,
|
| 2754 |
+
"step": 1950
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 1.0,
|
| 2758 |
+
"step": 1950,
|
| 2759 |
+
"total_flos": 204145164288000.0,
|
| 2760 |
+
"train_loss": 0.5252774189068721,
|
| 2761 |
+
"train_runtime": 19779.9981,
|
| 2762 |
+
"train_samples_per_second": 1.577,
|
| 2763 |
+
"train_steps_per_second": 0.099
|
| 2764 |
+
}
|
| 2765 |
+
],
|
| 2766 |
+
"logging_steps": 5,
|
| 2767 |
+
"max_steps": 1950,
|
| 2768 |
+
"num_input_tokens_seen": 0,
|
| 2769 |
+
"num_train_epochs": 1,
|
| 2770 |
+
"save_steps": 100,
|
| 2771 |
+
"stateful_callbacks": {
|
| 2772 |
+
"TrainerControl": {
|
| 2773 |
+
"args": {
|
| 2774 |
+
"should_epoch_stop": false,
|
| 2775 |
+
"should_evaluate": false,
|
| 2776 |
+
"should_log": false,
|
| 2777 |
+
"should_save": true,
|
| 2778 |
+
"should_training_stop": true
|
| 2779 |
+
},
|
| 2780 |
+
"attributes": {}
|
| 2781 |
+
}
|
| 2782 |
+
},
|
| 2783 |
+
"total_flos": 204145164288000.0,
|
| 2784 |
+
"train_batch_size": 4,
|
| 2785 |
+
"trial_name": null,
|
| 2786 |
+
"trial_params": null
|
| 2787 |
+
}
|