Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep24_06-30-26_COE-CS-sv003/events.out.tfevents.1727159729.COE-CS-sv003.774726.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2653 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: feedback_p0.1_seed42_level2_syntaxdirtybatch16_reverse
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# feedback_p0.1_seed42_level2_syntaxdirtybatch16_reverse
|
| 18 |
+
|
| 19 |
+
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 None dataset.
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 1e-05
|
| 39 |
+
- train_batch_size: 4
|
| 40 |
+
- eval_batch_size: 1
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 4
|
| 44 |
+
- total_train_batch_size: 16
|
| 45 |
+
- total_eval_batch_size: 4
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- Transformers 4.44.2
|
| 58 |
+
- Pytorch 2.3.1+cu121
|
| 59 |
+
- Datasets 2.19.1
|
| 60 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 194409010298880.0,
|
| 4 |
+
"train_loss": 0.5086988852623655,
|
| 5 |
+
"train_runtime": 16625.9643,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.787,
|
| 8 |
+
"train_steps_per_second": 0.112
|
| 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": "4.44.2"
|
| 6 |
+
}
|
runs/Sep24_06-30-26_COE-CS-sv003/events.out.tfevents.1727159729.COE-CS-sv003.774726.0
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:cfeba154f7ecaea31cfc46476458f13e44be295d24885f01358b04f55bf3d8ca
|
| 3 |
+
size 84271
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 194409010298880.0,
|
| 4 |
+
"train_loss": 0.5086988852623655,
|
| 5 |
+
"train_runtime": 16625.9643,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.787,
|
| 8 |
+
"train_steps_per_second": 0.112
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2653 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1857,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005385029617662897,
|
| 13 |
+
"grad_norm": 22.7613713160104,
|
| 14 |
+
"learning_rate": 5.376344086021506e-08,
|
| 15 |
+
"loss": 1.3472,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0026925148088314485,
|
| 20 |
+
"grad_norm": 22.37683485347051,
|
| 21 |
+
"learning_rate": 2.688172043010753e-07,
|
| 22 |
+
"loss": 1.334,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005385029617662897,
|
| 27 |
+
"grad_norm": 16.198717013287386,
|
| 28 |
+
"learning_rate": 5.376344086021506e-07,
|
| 29 |
+
"loss": 1.3098,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008077544426494346,
|
| 34 |
+
"grad_norm": 12.105460048613201,
|
| 35 |
+
"learning_rate": 8.064516129032258e-07,
|
| 36 |
+
"loss": 1.1643,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010770059235325794,
|
| 41 |
+
"grad_norm": 7.514844224376465,
|
| 42 |
+
"learning_rate": 1.0752688172043011e-06,
|
| 43 |
+
"loss": 1.0474,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013462574044157244,
|
| 48 |
+
"grad_norm": 3.8632002795574367,
|
| 49 |
+
"learning_rate": 1.3440860215053765e-06,
|
| 50 |
+
"loss": 0.9317,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01615508885298869,
|
| 55 |
+
"grad_norm": 3.400445342173989,
|
| 56 |
+
"learning_rate": 1.6129032258064516e-06,
|
| 57 |
+
"loss": 0.8826,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.01884760366182014,
|
| 62 |
+
"grad_norm": 2.9898366351100547,
|
| 63 |
+
"learning_rate": 1.881720430107527e-06,
|
| 64 |
+
"loss": 0.8635,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.021540118470651588,
|
| 69 |
+
"grad_norm": 2.8585969196090373,
|
| 70 |
+
"learning_rate": 2.1505376344086023e-06,
|
| 71 |
+
"loss": 0.8577,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.024232633279483037,
|
| 76 |
+
"grad_norm": 3.0122772432907277,
|
| 77 |
+
"learning_rate": 2.4193548387096776e-06,
|
| 78 |
+
"loss": 0.8418,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.026925148088314487,
|
| 83 |
+
"grad_norm": 3.0903177647257336,
|
| 84 |
+
"learning_rate": 2.688172043010753e-06,
|
| 85 |
+
"loss": 0.8311,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.029617662897145933,
|
| 90 |
+
"grad_norm": 3.0109701116495216,
|
| 91 |
+
"learning_rate": 2.9569892473118283e-06,
|
| 92 |
+
"loss": 0.8072,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03231017770597738,
|
| 97 |
+
"grad_norm": 2.9583394058653876,
|
| 98 |
+
"learning_rate": 3.225806451612903e-06,
|
| 99 |
+
"loss": 0.7965,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03500269251480883,
|
| 104 |
+
"grad_norm": 3.1460179996258146,
|
| 105 |
+
"learning_rate": 3.494623655913979e-06,
|
| 106 |
+
"loss": 0.8118,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03769520732364028,
|
| 111 |
+
"grad_norm": 3.376763980257778,
|
| 112 |
+
"learning_rate": 3.763440860215054e-06,
|
| 113 |
+
"loss": 0.7851,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04038772213247173,
|
| 118 |
+
"grad_norm": 2.9673425200438013,
|
| 119 |
+
"learning_rate": 4.032258064516129e-06,
|
| 120 |
+
"loss": 0.7957,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.043080236941303175,
|
| 125 |
+
"grad_norm": 3.0128704990376267,
|
| 126 |
+
"learning_rate": 4.3010752688172045e-06,
|
| 127 |
+
"loss": 0.7751,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04577275175013463,
|
| 132 |
+
"grad_norm": 3.3210132634127656,
|
| 133 |
+
"learning_rate": 4.56989247311828e-06,
|
| 134 |
+
"loss": 0.7689,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.048465266558966075,
|
| 139 |
+
"grad_norm": 3.175041603297429,
|
| 140 |
+
"learning_rate": 4.838709677419355e-06,
|
| 141 |
+
"loss": 0.7685,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05115778136779752,
|
| 146 |
+
"grad_norm": 3.1666439879487838,
|
| 147 |
+
"learning_rate": 5.1075268817204305e-06,
|
| 148 |
+
"loss": 0.7642,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.053850296176628974,
|
| 153 |
+
"grad_norm": 3.050039288877326,
|
| 154 |
+
"learning_rate": 5.376344086021506e-06,
|
| 155 |
+
"loss": 0.75,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05654281098546042,
|
| 160 |
+
"grad_norm": 2.9472819168277455,
|
| 161 |
+
"learning_rate": 5.645161290322582e-06,
|
| 162 |
+
"loss": 0.7437,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05923532579429187,
|
| 167 |
+
"grad_norm": 3.1420868913301185,
|
| 168 |
+
"learning_rate": 5.9139784946236566e-06,
|
| 169 |
+
"loss": 0.7317,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06192784060312332,
|
| 174 |
+
"grad_norm": 3.2485163047441423,
|
| 175 |
+
"learning_rate": 6.182795698924732e-06,
|
| 176 |
+
"loss": 0.7392,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06462035541195477,
|
| 181 |
+
"grad_norm": 3.4854700395087552,
|
| 182 |
+
"learning_rate": 6.451612903225806e-06,
|
| 183 |
+
"loss": 0.7441,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06731287022078622,
|
| 188 |
+
"grad_norm": 2.9119403460101148,
|
| 189 |
+
"learning_rate": 6.720430107526882e-06,
|
| 190 |
+
"loss": 0.7216,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07000538502961766,
|
| 195 |
+
"grad_norm": 2.956926563215588,
|
| 196 |
+
"learning_rate": 6.989247311827958e-06,
|
| 197 |
+
"loss": 0.7283,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07269789983844911,
|
| 202 |
+
"grad_norm": 3.256266278148656,
|
| 203 |
+
"learning_rate": 7.258064516129033e-06,
|
| 204 |
+
"loss": 0.7296,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07539041464728057,
|
| 209 |
+
"grad_norm": 2.8692943178965185,
|
| 210 |
+
"learning_rate": 7.526881720430108e-06,
|
| 211 |
+
"loss": 0.7202,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.078082929456112,
|
| 216 |
+
"grad_norm": 2.8804161059666935,
|
| 217 |
+
"learning_rate": 7.795698924731183e-06,
|
| 218 |
+
"loss": 0.7195,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08077544426494346,
|
| 223 |
+
"grad_norm": 3.0006510283335848,
|
| 224 |
+
"learning_rate": 8.064516129032258e-06,
|
| 225 |
+
"loss": 0.721,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08346795907377491,
|
| 230 |
+
"grad_norm": 2.978282671191466,
|
| 231 |
+
"learning_rate": 8.333333333333334e-06,
|
| 232 |
+
"loss": 0.7151,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08616047388260635,
|
| 237 |
+
"grad_norm": 2.832558314575903,
|
| 238 |
+
"learning_rate": 8.602150537634409e-06,
|
| 239 |
+
"loss": 0.712,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.0888529886914378,
|
| 244 |
+
"grad_norm": 2.9438520239067265,
|
| 245 |
+
"learning_rate": 8.870967741935484e-06,
|
| 246 |
+
"loss": 0.7161,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09154550350026926,
|
| 251 |
+
"grad_norm": 3.165458696824085,
|
| 252 |
+
"learning_rate": 9.13978494623656e-06,
|
| 253 |
+
"loss": 0.7146,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0942380183091007,
|
| 258 |
+
"grad_norm": 3.372895429211009,
|
| 259 |
+
"learning_rate": 9.408602150537635e-06,
|
| 260 |
+
"loss": 0.7008,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09693053311793215,
|
| 265 |
+
"grad_norm": 2.9822482771935785,
|
| 266 |
+
"learning_rate": 9.67741935483871e-06,
|
| 267 |
+
"loss": 0.702,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.0996230479267636,
|
| 272 |
+
"grad_norm": 2.7297175029131555,
|
| 273 |
+
"learning_rate": 9.946236559139786e-06,
|
| 274 |
+
"loss": 0.701,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10231556273559504,
|
| 279 |
+
"grad_norm": 2.797512257800779,
|
| 280 |
+
"learning_rate": 9.999858614526654e-06,
|
| 281 |
+
"loss": 0.7073,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1050080775444265,
|
| 286 |
+
"grad_norm": 2.7494639915648516,
|
| 287 |
+
"learning_rate": 9.999284249745143e-06,
|
| 288 |
+
"loss": 0.6922,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10770059235325795,
|
| 293 |
+
"grad_norm": 2.8999873055676404,
|
| 294 |
+
"learning_rate": 9.998268119778471e-06,
|
| 295 |
+
"loss": 0.6998,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.11039310716208939,
|
| 300 |
+
"grad_norm": 2.709033463381412,
|
| 301 |
+
"learning_rate": 9.996810314417666e-06,
|
| 302 |
+
"loss": 0.7017,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11308562197092084,
|
| 307 |
+
"grad_norm": 2.6497064978122498,
|
| 308 |
+
"learning_rate": 9.9949109624827e-06,
|
| 309 |
+
"loss": 0.697,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1157781367797523,
|
| 314 |
+
"grad_norm": 2.850561924743422,
|
| 315 |
+
"learning_rate": 9.992570231811108e-06,
|
| 316 |
+
"loss": 0.6937,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11847065158858373,
|
| 321 |
+
"grad_norm": 2.7049716604581953,
|
| 322 |
+
"learning_rate": 9.989788329243163e-06,
|
| 323 |
+
"loss": 0.7057,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.12116316639741519,
|
| 328 |
+
"grad_norm": 2.5917851309791673,
|
| 329 |
+
"learning_rate": 9.986565500603598e-06,
|
| 330 |
+
"loss": 0.6956,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12385568120624664,
|
| 335 |
+
"grad_norm": 2.6711850256792258,
|
| 336 |
+
"learning_rate": 9.982902030679875e-06,
|
| 337 |
+
"loss": 0.7109,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12654819601507808,
|
| 342 |
+
"grad_norm": 2.6766440274545995,
|
| 343 |
+
"learning_rate": 9.978798243197031e-06,
|
| 344 |
+
"loss": 0.6909,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12924071082390953,
|
| 349 |
+
"grad_norm": 2.8552384364271775,
|
| 350 |
+
"learning_rate": 9.974254500789065e-06,
|
| 351 |
+
"loss": 0.6879,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.13193322563274099,
|
| 356 |
+
"grad_norm": 2.5756385146469882,
|
| 357 |
+
"learning_rate": 9.969271204966895e-06,
|
| 358 |
+
"loss": 0.6802,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13462574044157244,
|
| 363 |
+
"grad_norm": 2.717537618462663,
|
| 364 |
+
"learning_rate": 9.963848796082873e-06,
|
| 365 |
+
"loss": 0.6921,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13731825525040386,
|
| 370 |
+
"grad_norm": 2.5131980662996756,
|
| 371 |
+
"learning_rate": 9.95798775329189e-06,
|
| 372 |
+
"loss": 0.689,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.14001077005923532,
|
| 377 |
+
"grad_norm": 2.51836898271677,
|
| 378 |
+
"learning_rate": 9.951688594509012e-06,
|
| 379 |
+
"loss": 0.6793,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14270328486806677,
|
| 384 |
+
"grad_norm": 2.6980273572649676,
|
| 385 |
+
"learning_rate": 9.944951876363731e-06,
|
| 386 |
+
"loss": 0.6834,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14539579967689822,
|
| 391 |
+
"grad_norm": 2.646134452091696,
|
| 392 |
+
"learning_rate": 9.937778194150771e-06,
|
| 393 |
+
"loss": 0.6767,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14808831448572968,
|
| 398 |
+
"grad_norm": 2.5534608799808627,
|
| 399 |
+
"learning_rate": 9.930168181777486e-06,
|
| 400 |
+
"loss": 0.6907,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.15078082929456113,
|
| 405 |
+
"grad_norm": 2.5908475307902656,
|
| 406 |
+
"learning_rate": 9.922122511707843e-06,
|
| 407 |
+
"loss": 0.6861,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.15347334410339256,
|
| 412 |
+
"grad_norm": 2.5765130409804695,
|
| 413 |
+
"learning_rate": 9.913641894903006e-06,
|
| 414 |
+
"loss": 0.6791,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.156165858912224,
|
| 419 |
+
"grad_norm": 2.741146958946548,
|
| 420 |
+
"learning_rate": 9.904727080758495e-06,
|
| 421 |
+
"loss": 0.6874,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15885837372105546,
|
| 426 |
+
"grad_norm": 2.8381749428470595,
|
| 427 |
+
"learning_rate": 9.895378857037989e-06,
|
| 428 |
+
"loss": 0.676,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.16155088852988692,
|
| 433 |
+
"grad_norm": 2.7204348331914447,
|
| 434 |
+
"learning_rate": 9.885598049803693e-06,
|
| 435 |
+
"loss": 0.673,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.16424340333871837,
|
| 440 |
+
"grad_norm": 2.6441432061304178,
|
| 441 |
+
"learning_rate": 9.875385523343353e-06,
|
| 442 |
+
"loss": 0.671,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16693591814754982,
|
| 447 |
+
"grad_norm": 2.5486118216678277,
|
| 448 |
+
"learning_rate": 9.86474218009388e-06,
|
| 449 |
+
"loss": 0.6622,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16962843295638125,
|
| 454 |
+
"grad_norm": 2.448370238657585,
|
| 455 |
+
"learning_rate": 9.853668960561611e-06,
|
| 456 |
+
"loss": 0.6848,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1723209477652127,
|
| 461 |
+
"grad_norm": 2.5840299111800955,
|
| 462 |
+
"learning_rate": 9.84216684323919e-06,
|
| 463 |
+
"loss": 0.6568,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17501346257404415,
|
| 468 |
+
"grad_norm": 2.57305271825157,
|
| 469 |
+
"learning_rate": 9.830236844519114e-06,
|
| 470 |
+
"loss": 0.66,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.1777059773828756,
|
| 475 |
+
"grad_norm": 2.53266567006045,
|
| 476 |
+
"learning_rate": 9.81788001860391e-06,
|
| 477 |
+
"loss": 0.662,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.18039849219170706,
|
| 482 |
+
"grad_norm": 2.4978159844537458,
|
| 483 |
+
"learning_rate": 9.805097457412981e-06,
|
| 484 |
+
"loss": 0.6509,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.1830910070005385,
|
| 489 |
+
"grad_norm": 2.533762872533261,
|
| 490 |
+
"learning_rate": 9.791890290486127e-06,
|
| 491 |
+
"loss": 0.6336,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18578352180936994,
|
| 496 |
+
"grad_norm": 2.5709499517182888,
|
| 497 |
+
"learning_rate": 9.77825968488372e-06,
|
| 498 |
+
"loss": 0.6672,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1884760366182014,
|
| 503 |
+
"grad_norm": 2.5155435537189432,
|
| 504 |
+
"learning_rate": 9.76420684508358e-06,
|
| 505 |
+
"loss": 0.6587,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.19116855142703285,
|
| 510 |
+
"grad_norm": 2.3629608947394214,
|
| 511 |
+
"learning_rate": 9.74973301287455e-06,
|
| 512 |
+
"loss": 0.6382,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.1938610662358643,
|
| 517 |
+
"grad_norm": 2.423886375936418,
|
| 518 |
+
"learning_rate": 9.734839467246744e-06,
|
| 519 |
+
"loss": 0.6399,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19655358104469575,
|
| 524 |
+
"grad_norm": 2.4151592067478713,
|
| 525 |
+
"learning_rate": 9.719527524278552e-06,
|
| 526 |
+
"loss": 0.6567,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1992460958535272,
|
| 531 |
+
"grad_norm": 2.339311457038808,
|
| 532 |
+
"learning_rate": 9.703798537020327e-06,
|
| 533 |
+
"loss": 0.6329,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.20193861066235863,
|
| 538 |
+
"grad_norm": 2.401955911603551,
|
| 539 |
+
"learning_rate": 9.687653895374824e-06,
|
| 540 |
+
"loss": 0.6443,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.20463112547119008,
|
| 545 |
+
"grad_norm": 2.349530919733858,
|
| 546 |
+
"learning_rate": 9.671095025974388e-06,
|
| 547 |
+
"loss": 0.6262,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20732364028002154,
|
| 552 |
+
"grad_norm": 2.5574771679537,
|
| 553 |
+
"learning_rate": 9.654123392054884e-06,
|
| 554 |
+
"loss": 0.6493,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.210016155088853,
|
| 559 |
+
"grad_norm": 2.5279142423497443,
|
| 560 |
+
"learning_rate": 9.636740493326398e-06,
|
| 561 |
+
"loss": 0.6458,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.21270866989768444,
|
| 566 |
+
"grad_norm": 2.407565936871044,
|
| 567 |
+
"learning_rate": 9.618947865840705e-06,
|
| 568 |
+
"loss": 0.6356,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2154011847065159,
|
| 573 |
+
"grad_norm": 2.50737386034651,
|
| 574 |
+
"learning_rate": 9.600747081855553e-06,
|
| 575 |
+
"loss": 0.6505,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.21809369951534732,
|
| 580 |
+
"grad_norm": 2.492501588167292,
|
| 581 |
+
"learning_rate": 9.582139749695713e-06,
|
| 582 |
+
"loss": 0.625,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.22078621432417878,
|
| 587 |
+
"grad_norm": 2.3085166355816322,
|
| 588 |
+
"learning_rate": 9.563127513610868e-06,
|
| 589 |
+
"loss": 0.6531,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.22347872913301023,
|
| 594 |
+
"grad_norm": 2.3663625111521553,
|
| 595 |
+
"learning_rate": 9.543712053630314e-06,
|
| 596 |
+
"loss": 0.6185,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22617124394184168,
|
| 601 |
+
"grad_norm": 2.362289550568963,
|
| 602 |
+
"learning_rate": 9.523895085414503e-06,
|
| 603 |
+
"loss": 0.6151,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22886375875067314,
|
| 608 |
+
"grad_norm": 2.4285748744254394,
|
| 609 |
+
"learning_rate": 9.503678360103433e-06,
|
| 610 |
+
"loss": 0.6195,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2315562735595046,
|
| 615 |
+
"grad_norm": 2.320249188264771,
|
| 616 |
+
"learning_rate": 9.483063664161922e-06,
|
| 617 |
+
"loss": 0.6217,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.23424878836833601,
|
| 622 |
+
"grad_norm": 2.4068743254230953,
|
| 623 |
+
"learning_rate": 9.462052819221726e-06,
|
| 624 |
+
"loss": 0.6269,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23694130317716747,
|
| 629 |
+
"grad_norm": 2.443692426314357,
|
| 630 |
+
"learning_rate": 9.440647681920587e-06,
|
| 631 |
+
"loss": 0.6414,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23963381798599892,
|
| 636 |
+
"grad_norm": 2.376216504990232,
|
| 637 |
+
"learning_rate": 9.418850143738158e-06,
|
| 638 |
+
"loss": 0.6349,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.24232633279483037,
|
| 643 |
+
"grad_norm": 2.4206153603989566,
|
| 644 |
+
"learning_rate": 9.396662130828869e-06,
|
| 645 |
+
"loss": 0.6059,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.24501884760366183,
|
| 650 |
+
"grad_norm": 2.4419798835688074,
|
| 651 |
+
"learning_rate": 9.374085603851709e-06,
|
| 652 |
+
"loss": 0.6091,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24771136241249328,
|
| 657 |
+
"grad_norm": 2.3263506335508817,
|
| 658 |
+
"learning_rate": 9.351122557796987e-06,
|
| 659 |
+
"loss": 0.6188,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.25040387722132473,
|
| 664 |
+
"grad_norm": 2.533382037846899,
|
| 665 |
+
"learning_rate": 9.327775021810037e-06,
|
| 666 |
+
"loss": 0.6145,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.25309639203015616,
|
| 671 |
+
"grad_norm": 2.678311932146126,
|
| 672 |
+
"learning_rate": 9.304045059011905e-06,
|
| 673 |
+
"loss": 0.5881,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.25578890683898764,
|
| 678 |
+
"grad_norm": 2.5127273745887466,
|
| 679 |
+
"learning_rate": 9.279934766317045e-06,
|
| 680 |
+
"loss": 0.621,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.25848142164781907,
|
| 685 |
+
"grad_norm": 2.303317205670643,
|
| 686 |
+
"learning_rate": 9.255446274248023e-06,
|
| 687 |
+
"loss": 0.6036,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2611739364566505,
|
| 692 |
+
"grad_norm": 2.4449467721888376,
|
| 693 |
+
"learning_rate": 9.230581746747257e-06,
|
| 694 |
+
"loss": 0.6035,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.26386645126548197,
|
| 699 |
+
"grad_norm": 2.4020663292502165,
|
| 700 |
+
"learning_rate": 9.205343380985793e-06,
|
| 701 |
+
"loss": 0.5939,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2665589660743134,
|
| 706 |
+
"grad_norm": 2.526561720865294,
|
| 707 |
+
"learning_rate": 9.179733407169146e-06,
|
| 708 |
+
"loss": 0.6,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2692514808831449,
|
| 713 |
+
"grad_norm": 2.4642667203514783,
|
| 714 |
+
"learning_rate": 9.153754088340237e-06,
|
| 715 |
+
"loss": 0.6143,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2719439956919763,
|
| 720 |
+
"grad_norm": 2.546297508506232,
|
| 721 |
+
"learning_rate": 9.12740772017941e-06,
|
| 722 |
+
"loss": 0.6006,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.27463651050080773,
|
| 727 |
+
"grad_norm": 2.332663269345773,
|
| 728 |
+
"learning_rate": 9.10069663080158e-06,
|
| 729 |
+
"loss": 0.5894,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2773290253096392,
|
| 734 |
+
"grad_norm": 2.414500443877397,
|
| 735 |
+
"learning_rate": 9.073623180550496e-06,
|
| 736 |
+
"loss": 0.5978,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.28002154011847064,
|
| 741 |
+
"grad_norm": 2.618196788357666,
|
| 742 |
+
"learning_rate": 9.046189761790179e-06,
|
| 743 |
+
"loss": 0.5914,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.2827140549273021,
|
| 748 |
+
"grad_norm": 2.4760448307330964,
|
| 749 |
+
"learning_rate": 9.018398798693512e-06,
|
| 750 |
+
"loss": 0.6094,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.28540656973613354,
|
| 755 |
+
"grad_norm": 2.403492095012741,
|
| 756 |
+
"learning_rate": 8.99025274702803e-06,
|
| 757 |
+
"loss": 0.5792,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.288099084544965,
|
| 762 |
+
"grad_norm": 2.424533931438387,
|
| 763 |
+
"learning_rate": 8.961754093938917e-06,
|
| 764 |
+
"loss": 0.5813,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.29079159935379645,
|
| 769 |
+
"grad_norm": 2.4025980246295555,
|
| 770 |
+
"learning_rate": 8.932905357729213e-06,
|
| 771 |
+
"loss": 0.5747,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2934841141626279,
|
| 776 |
+
"grad_norm": 2.4566763739595343,
|
| 777 |
+
"learning_rate": 8.903709087637307e-06,
|
| 778 |
+
"loss": 0.6003,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.29617662897145935,
|
| 783 |
+
"grad_norm": 2.307834128807205,
|
| 784 |
+
"learning_rate": 8.874167863611647e-06,
|
| 785 |
+
"loss": 0.5838,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.2988691437802908,
|
| 790 |
+
"grad_norm": 2.352469567943909,
|
| 791 |
+
"learning_rate": 8.844284296082776e-06,
|
| 792 |
+
"loss": 0.5845,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.30156165858912226,
|
| 797 |
+
"grad_norm": 2.3341023442033895,
|
| 798 |
+
"learning_rate": 8.81406102573266e-06,
|
| 799 |
+
"loss": 0.5689,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3042541733979537,
|
| 804 |
+
"grad_norm": 2.4151388881341322,
|
| 805 |
+
"learning_rate": 8.783500723261333e-06,
|
| 806 |
+
"loss": 0.5793,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3069466882067851,
|
| 811 |
+
"grad_norm": 2.352881741468682,
|
| 812 |
+
"learning_rate": 8.752606089150903e-06,
|
| 813 |
+
"loss": 0.5872,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3096392030156166,
|
| 818 |
+
"grad_norm": 2.2929845371154665,
|
| 819 |
+
"learning_rate": 8.721379853426928e-06,
|
| 820 |
+
"loss": 0.5827,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.312331717824448,
|
| 825 |
+
"grad_norm": 2.351649375339974,
|
| 826 |
+
"learning_rate": 8.689824775417164e-06,
|
| 827 |
+
"loss": 0.5704,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3150242326332795,
|
| 832 |
+
"grad_norm": 2.3536783488717425,
|
| 833 |
+
"learning_rate": 8.657943643507747e-06,
|
| 834 |
+
"loss": 0.5916,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3177167474421109,
|
| 839 |
+
"grad_norm": 2.4858567387352783,
|
| 840 |
+
"learning_rate": 8.625739274896785e-06,
|
| 841 |
+
"loss": 0.5679,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3204092622509424,
|
| 846 |
+
"grad_norm": 2.5420824976461613,
|
| 847 |
+
"learning_rate": 8.593214515345426e-06,
|
| 848 |
+
"loss": 0.5747,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.32310177705977383,
|
| 853 |
+
"grad_norm": 2.411084856535756,
|
| 854 |
+
"learning_rate": 8.560372238926372e-06,
|
| 855 |
+
"loss": 0.5818,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.32579429186860526,
|
| 860 |
+
"grad_norm": 2.493451871387303,
|
| 861 |
+
"learning_rate": 8.527215347769932e-06,
|
| 862 |
+
"loss": 0.5709,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.32848680667743674,
|
| 867 |
+
"grad_norm": 2.2974929968587388,
|
| 868 |
+
"learning_rate": 8.493746771807559e-06,
|
| 869 |
+
"loss": 0.578,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.33117932148626816,
|
| 874 |
+
"grad_norm": 2.368635502736958,
|
| 875 |
+
"learning_rate": 8.459969468512943e-06,
|
| 876 |
+
"loss": 0.5507,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.33387183629509964,
|
| 881 |
+
"grad_norm": 2.5093780273624384,
|
| 882 |
+
"learning_rate": 8.425886422640684e-06,
|
| 883 |
+
"loss": 0.5574,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.33656435110393107,
|
| 888 |
+
"grad_norm": 2.3784956807998254,
|
| 889 |
+
"learning_rate": 8.391500645962527e-06,
|
| 890 |
+
"loss": 0.5629,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3392568659127625,
|
| 895 |
+
"grad_norm": 2.5705211303648987,
|
| 896 |
+
"learning_rate": 8.356815177001243e-06,
|
| 897 |
+
"loss": 0.5563,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.341949380721594,
|
| 902 |
+
"grad_norm": 2.296265064615366,
|
| 903 |
+
"learning_rate": 8.321833080762109e-06,
|
| 904 |
+
"loss": 0.5673,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3446418955304254,
|
| 909 |
+
"grad_norm": 2.468686941019328,
|
| 910 |
+
"learning_rate": 8.286557448462074e-06,
|
| 911 |
+
"loss": 0.5615,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3473344103392569,
|
| 916 |
+
"grad_norm": 2.304658882065399,
|
| 917 |
+
"learning_rate": 8.250991397256609e-06,
|
| 918 |
+
"loss": 0.5716,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3500269251480883,
|
| 923 |
+
"grad_norm": 2.4328147271748968,
|
| 924 |
+
"learning_rate": 8.215138069964248e-06,
|
| 925 |
+
"loss": 0.5497,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3527194399569198,
|
| 930 |
+
"grad_norm": 2.5638444164193626,
|
| 931 |
+
"learning_rate": 8.179000634788872e-06,
|
| 932 |
+
"loss": 0.5634,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3554119547657512,
|
| 937 |
+
"grad_norm": 2.304000568970403,
|
| 938 |
+
"learning_rate": 8.142582285039753e-06,
|
| 939 |
+
"loss": 0.5552,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.35810446957458264,
|
| 944 |
+
"grad_norm": 2.312118564540539,
|
| 945 |
+
"learning_rate": 8.105886238849369e-06,
|
| 946 |
+
"loss": 0.5665,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3607969843834141,
|
| 951 |
+
"grad_norm": 2.564926504630545,
|
| 952 |
+
"learning_rate": 8.068915738889038e-06,
|
| 953 |
+
"loss": 0.5432,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.36348949919224555,
|
| 958 |
+
"grad_norm": 2.552416968513296,
|
| 959 |
+
"learning_rate": 8.031674052082372e-06,
|
| 960 |
+
"loss": 0.5418,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.366182014001077,
|
| 965 |
+
"grad_norm": 2.465276318426562,
|
| 966 |
+
"learning_rate": 7.994164469316596e-06,
|
| 967 |
+
"loss": 0.5438,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.36887452880990845,
|
| 972 |
+
"grad_norm": 2.3946509022495728,
|
| 973 |
+
"learning_rate": 7.956390305151744e-06,
|
| 974 |
+
"loss": 0.5399,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3715670436187399,
|
| 979 |
+
"grad_norm": 2.5006069528524266,
|
| 980 |
+
"learning_rate": 7.918354897527767e-06,
|
| 981 |
+
"loss": 0.5556,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.37425955842757136,
|
| 986 |
+
"grad_norm": 2.4646542978549633,
|
| 987 |
+
"learning_rate": 7.880061607469574e-06,
|
| 988 |
+
"loss": 0.5465,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3769520732364028,
|
| 993 |
+
"grad_norm": 2.464786801902601,
|
| 994 |
+
"learning_rate": 7.841513818790035e-06,
|
| 995 |
+
"loss": 0.5412,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.37964458804523427,
|
| 1000 |
+
"grad_norm": 2.470119700617401,
|
| 1001 |
+
"learning_rate": 7.80271493779096e-06,
|
| 1002 |
+
"loss": 0.5518,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3823371028540657,
|
| 1007 |
+
"grad_norm": 2.487807822690958,
|
| 1008 |
+
"learning_rate": 7.763668392962107e-06,
|
| 1009 |
+
"loss": 0.539,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3850296176628972,
|
| 1014 |
+
"grad_norm": 2.3449768254321812,
|
| 1015 |
+
"learning_rate": 7.724377634678222e-06,
|
| 1016 |
+
"loss": 0.5367,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.3877221324717286,
|
| 1021 |
+
"grad_norm": 2.342137329316273,
|
| 1022 |
+
"learning_rate": 7.684846134894133e-06,
|
| 1023 |
+
"loss": 0.5361,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.39041464728056,
|
| 1028 |
+
"grad_norm": 2.3478632374387236,
|
| 1029 |
+
"learning_rate": 7.645077386837963e-06,
|
| 1030 |
+
"loss": 0.5377,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.3931071620893915,
|
| 1035 |
+
"grad_norm": 2.311127470477535,
|
| 1036 |
+
"learning_rate": 7.605074904702438e-06,
|
| 1037 |
+
"loss": 0.5326,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.39579967689822293,
|
| 1042 |
+
"grad_norm": 2.5257265487257317,
|
| 1043 |
+
"learning_rate": 7.5648422233343564e-06,
|
| 1044 |
+
"loss": 0.5253,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3984921917070544,
|
| 1049 |
+
"grad_norm": 2.2883296027896862,
|
| 1050 |
+
"learning_rate": 7.524382897922229e-06,
|
| 1051 |
+
"loss": 0.5198,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.40118470651588584,
|
| 1056 |
+
"grad_norm": 2.2938224585559994,
|
| 1057 |
+
"learning_rate": 7.483700503682127e-06,
|
| 1058 |
+
"loss": 0.5179,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.40387722132471726,
|
| 1063 |
+
"grad_norm": 2.351764729142054,
|
| 1064 |
+
"learning_rate": 7.442798635541749e-06,
|
| 1065 |
+
"loss": 0.5245,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.40656973613354874,
|
| 1070 |
+
"grad_norm": 2.2446519389708057,
|
| 1071 |
+
"learning_rate": 7.401680907822756e-06,
|
| 1072 |
+
"loss": 0.5356,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.40926225094238017,
|
| 1077 |
+
"grad_norm": 2.463243359027525,
|
| 1078 |
+
"learning_rate": 7.360350953921386e-06,
|
| 1079 |
+
"loss": 0.5261,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.41195476575121165,
|
| 1084 |
+
"grad_norm": 2.1930439197029634,
|
| 1085 |
+
"learning_rate": 7.3188124259873946e-06,
|
| 1086 |
+
"loss": 0.5401,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4146472805600431,
|
| 1091 |
+
"grad_norm": 2.37252598970325,
|
| 1092 |
+
"learning_rate": 7.2770689946013215e-06,
|
| 1093 |
+
"loss": 0.5047,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.41733979536887456,
|
| 1098 |
+
"grad_norm": 2.284338792322131,
|
| 1099 |
+
"learning_rate": 7.235124348450143e-06,
|
| 1100 |
+
"loss": 0.5121,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.420032310177706,
|
| 1105 |
+
"grad_norm": 2.253488965644508,
|
| 1106 |
+
"learning_rate": 7.192982194001312e-06,
|
| 1107 |
+
"loss": 0.5029,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4227248249865374,
|
| 1112 |
+
"grad_norm": 2.258949760855827,
|
| 1113 |
+
"learning_rate": 7.150646255175242e-06,
|
| 1114 |
+
"loss": 0.5282,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4254173397953689,
|
| 1119 |
+
"grad_norm": 2.2735662795523317,
|
| 1120 |
+
"learning_rate": 7.108120273016236e-06,
|
| 1121 |
+
"loss": 0.514,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4281098546042003,
|
| 1126 |
+
"grad_norm": 2.2850305304777865,
|
| 1127 |
+
"learning_rate": 7.065408005361902e-06,
|
| 1128 |
+
"loss": 0.5046,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4308023694130318,
|
| 1133 |
+
"grad_norm": 2.2431623733294925,
|
| 1134 |
+
"learning_rate": 7.022513226511098e-06,
|
| 1135 |
+
"loss": 0.5108,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4334948842218632,
|
| 1140 |
+
"grad_norm": 2.3702355406435958,
|
| 1141 |
+
"learning_rate": 6.9794397268904094e-06,
|
| 1142 |
+
"loss": 0.5039,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.43618739903069464,
|
| 1147 |
+
"grad_norm": 2.424641619769436,
|
| 1148 |
+
"learning_rate": 6.9361913127192026e-06,
|
| 1149 |
+
"loss": 0.5135,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4388799138395261,
|
| 1154 |
+
"grad_norm": 2.417084290942502,
|
| 1155 |
+
"learning_rate": 6.892771805673289e-06,
|
| 1156 |
+
"loss": 0.5008,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.44157242864835755,
|
| 1161 |
+
"grad_norm": 2.483155905595343,
|
| 1162 |
+
"learning_rate": 6.8491850425472216e-06,
|
| 1163 |
+
"loss": 0.5061,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.44426494345718903,
|
| 1168 |
+
"grad_norm": 2.3647966120276624,
|
| 1169 |
+
"learning_rate": 6.805434874915249e-06,
|
| 1170 |
+
"loss": 0.5116,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.44695745826602046,
|
| 1175 |
+
"grad_norm": 2.3941067808468235,
|
| 1176 |
+
"learning_rate": 6.7615251687909714e-06,
|
| 1177 |
+
"loss": 0.5115,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.44964997307485194,
|
| 1182 |
+
"grad_norm": 2.431013960473848,
|
| 1183 |
+
"learning_rate": 6.7174598042857186e-06,
|
| 1184 |
+
"loss": 0.4884,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.45234248788368336,
|
| 1189 |
+
"grad_norm": 2.3101487072378575,
|
| 1190 |
+
"learning_rate": 6.67324267526568e-06,
|
| 1191 |
+
"loss": 0.4952,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.4550350026925148,
|
| 1196 |
+
"grad_norm": 2.252689751264616,
|
| 1197 |
+
"learning_rate": 6.62887768900782e-06,
|
| 1198 |
+
"loss": 0.4908,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.45772751750134627,
|
| 1203 |
+
"grad_norm": 2.308986864251815,
|
| 1204 |
+
"learning_rate": 6.58436876585461e-06,
|
| 1205 |
+
"loss": 0.5115,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.4604200323101777,
|
| 1210 |
+
"grad_norm": 2.3582282125809333,
|
| 1211 |
+
"learning_rate": 6.539719838867604e-06,
|
| 1212 |
+
"loss": 0.5037,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4631125471190092,
|
| 1217 |
+
"grad_norm": 2.7549506575377065,
|
| 1218 |
+
"learning_rate": 6.494934853479889e-06,
|
| 1219 |
+
"loss": 0.4939,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4658050619278406,
|
| 1224 |
+
"grad_norm": 3.0601934459715916,
|
| 1225 |
+
"learning_rate": 6.450017767147447e-06,
|
| 1226 |
+
"loss": 0.4988,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.46849757673667203,
|
| 1231 |
+
"grad_norm": 2.3645028092509914,
|
| 1232 |
+
"learning_rate": 6.404972548999453e-06,
|
| 1233 |
+
"loss": 0.4962,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.4711900915455035,
|
| 1238 |
+
"grad_norm": 2.345655002717708,
|
| 1239 |
+
"learning_rate": 6.3598031794875355e-06,
|
| 1240 |
+
"loss": 0.4993,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.47388260635433493,
|
| 1245 |
+
"grad_norm": 2.4224055568308858,
|
| 1246 |
+
"learning_rate": 6.314513650034044e-06,
|
| 1247 |
+
"loss": 0.4798,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4765751211631664,
|
| 1252 |
+
"grad_norm": 2.3397132476622047,
|
| 1253 |
+
"learning_rate": 6.269107962679344e-06,
|
| 1254 |
+
"loss": 0.4879,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.47926763597199784,
|
| 1259 |
+
"grad_norm": 2.4522897916842914,
|
| 1260 |
+
"learning_rate": 6.223590129728171e-06,
|
| 1261 |
+
"loss": 0.4765,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4819601507808293,
|
| 1266 |
+
"grad_norm": 2.3108192463540593,
|
| 1267 |
+
"learning_rate": 6.177964173395087e-06,
|
| 1268 |
+
"loss": 0.4864,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.48465266558966075,
|
| 1273 |
+
"grad_norm": 2.267641247075609,
|
| 1274 |
+
"learning_rate": 6.1322341254490515e-06,
|
| 1275 |
+
"loss": 0.4836,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4873451803984922,
|
| 1280 |
+
"grad_norm": 2.2555810036335497,
|
| 1281 |
+
"learning_rate": 6.086404026857146e-06,
|
| 1282 |
+
"loss": 0.4842,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.49003769520732365,
|
| 1287 |
+
"grad_norm": 2.327632792501213,
|
| 1288 |
+
"learning_rate": 6.040477927427502e-06,
|
| 1289 |
+
"loss": 0.4704,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4927302100161551,
|
| 1294 |
+
"grad_norm": 2.2698670809887185,
|
| 1295 |
+
"learning_rate": 5.994459885451423e-06,
|
| 1296 |
+
"loss": 0.4749,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.49542272482498656,
|
| 1301 |
+
"grad_norm": 2.4725183020536847,
|
| 1302 |
+
"learning_rate": 5.948353967344786e-06,
|
| 1303 |
+
"loss": 0.4713,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.498115239633818,
|
| 1308 |
+
"grad_norm": 2.3182514849471993,
|
| 1309 |
+
"learning_rate": 5.902164247288701e-06,
|
| 1310 |
+
"loss": 0.4591,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5008077544426495,
|
| 1315 |
+
"grad_norm": 2.2450126172776543,
|
| 1316 |
+
"learning_rate": 5.855894806869493e-06,
|
| 1317 |
+
"loss": 0.4706,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5035002692514808,
|
| 1322 |
+
"grad_norm": 2.4095070380073333,
|
| 1323 |
+
"learning_rate": 5.809549734718031e-06,
|
| 1324 |
+
"loss": 0.4706,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5061927840603123,
|
| 1329 |
+
"grad_norm": 2.2933657857094905,
|
| 1330 |
+
"learning_rate": 5.763133126148438e-06,
|
| 1331 |
+
"loss": 0.4665,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5088852988691438,
|
| 1336 |
+
"grad_norm": 2.3063316226503123,
|
| 1337 |
+
"learning_rate": 5.716649082796199e-06,
|
| 1338 |
+
"loss": 0.4625,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5115778136779753,
|
| 1343 |
+
"grad_norm": 2.3233885177577966,
|
| 1344 |
+
"learning_rate": 5.670101712255716e-06,
|
| 1345 |
+
"loss": 0.4688,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5142703284868066,
|
| 1350 |
+
"grad_norm": 2.211725156600061,
|
| 1351 |
+
"learning_rate": 5.6234951277173524e-06,
|
| 1352 |
+
"loss": 0.4543,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5169628432956381,
|
| 1357 |
+
"grad_norm": 2.4389466884168147,
|
| 1358 |
+
"learning_rate": 5.576833447603943e-06,
|
| 1359 |
+
"loss": 0.4703,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5196553581044696,
|
| 1364 |
+
"grad_norm": 2.3368172021818725,
|
| 1365 |
+
"learning_rate": 5.530120795206894e-06,
|
| 1366 |
+
"loss": 0.4655,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.522347872913301,
|
| 1371 |
+
"grad_norm": 2.2744905601737804,
|
| 1372 |
+
"learning_rate": 5.483361298321803e-06,
|
| 1373 |
+
"loss": 0.4598,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5250403877221325,
|
| 1378 |
+
"grad_norm": 2.2417520684111354,
|
| 1379 |
+
"learning_rate": 5.4365590888837175e-06,
|
| 1380 |
+
"loss": 0.4542,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5277329025309639,
|
| 1385 |
+
"grad_norm": 2.260056934844272,
|
| 1386 |
+
"learning_rate": 5.389718302602008e-06,
|
| 1387 |
+
"loss": 0.4786,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5304254173397954,
|
| 1392 |
+
"grad_norm": 2.2937595883241,
|
| 1393 |
+
"learning_rate": 5.342843078594916e-06,
|
| 1394 |
+
"loss": 0.4649,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5331179321486268,
|
| 1399 |
+
"grad_norm": 2.298350623188229,
|
| 1400 |
+
"learning_rate": 5.295937559023794e-06,
|
| 1401 |
+
"loss": 0.4605,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5358104469574583,
|
| 1406 |
+
"grad_norm": 2.2861153437802915,
|
| 1407 |
+
"learning_rate": 5.2490058887270815e-06,
|
| 1408 |
+
"loss": 0.4529,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5385029617662898,
|
| 1413 |
+
"grad_norm": 2.2661516039174288,
|
| 1414 |
+
"learning_rate": 5.202052214854048e-06,
|
| 1415 |
+
"loss": 0.448,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5411954765751211,
|
| 1420 |
+
"grad_norm": 2.287756336798117,
|
| 1421 |
+
"learning_rate": 5.1550806864983235e-06,
|
| 1422 |
+
"loss": 0.4577,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5438879913839526,
|
| 1427 |
+
"grad_norm": 2.21952469191807,
|
| 1428 |
+
"learning_rate": 5.108095454331261e-06,
|
| 1429 |
+
"loss": 0.4401,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5465805061927841,
|
| 1434 |
+
"grad_norm": 2.289575301200996,
|
| 1435 |
+
"learning_rate": 5.061100670235161e-06,
|
| 1436 |
+
"loss": 0.4504,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5492730210016155,
|
| 1441 |
+
"grad_norm": 2.254586229300791,
|
| 1442 |
+
"learning_rate": 5.014100486936383e-06,
|
| 1443 |
+
"loss": 0.4454,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5519655358104469,
|
| 1448 |
+
"grad_norm": 2.137160999612689,
|
| 1449 |
+
"learning_rate": 4.967099057638397e-06,
|
| 1450 |
+
"loss": 0.4366,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5546580506192784,
|
| 1455 |
+
"grad_norm": 2.241158588636164,
|
| 1456 |
+
"learning_rate": 4.92010053565477e-06,
|
| 1457 |
+
"loss": 0.442,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5573505654281099,
|
| 1462 |
+
"grad_norm": 2.218608478823859,
|
| 1463 |
+
"learning_rate": 4.873109074042169e-06,
|
| 1464 |
+
"loss": 0.4346,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5600430802369413,
|
| 1469 |
+
"grad_norm": 2.43524119080364,
|
| 1470 |
+
"learning_rate": 4.826128825233362e-06,
|
| 1471 |
+
"loss": 0.4428,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5627355950457728,
|
| 1476 |
+
"grad_norm": 2.353557854902857,
|
| 1477 |
+
"learning_rate": 4.77916394067029e-06,
|
| 1478 |
+
"loss": 0.4414,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5654281098546042,
|
| 1483 |
+
"grad_norm": 2.2938687493350423,
|
| 1484 |
+
"learning_rate": 4.732218570437224e-06,
|
| 1485 |
+
"loss": 0.4568,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5681206246634356,
|
| 1490 |
+
"grad_norm": 2.3216946351133405,
|
| 1491 |
+
"learning_rate": 4.6852968628940345e-06,
|
| 1492 |
+
"loss": 0.4397,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5708131394722671,
|
| 1497 |
+
"grad_norm": 2.392290169077891,
|
| 1498 |
+
"learning_rate": 4.638402964309626e-06,
|
| 1499 |
+
"loss": 0.4323,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5735056542810986,
|
| 1504 |
+
"grad_norm": 2.259967562074028,
|
| 1505 |
+
"learning_rate": 4.591541018495543e-06,
|
| 1506 |
+
"loss": 0.4427,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.57619816908993,
|
| 1511 |
+
"grad_norm": 2.4321895943132597,
|
| 1512 |
+
"learning_rate": 4.5447151664398e-06,
|
| 1513 |
+
"loss": 0.44,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5788906838987614,
|
| 1518 |
+
"grad_norm": 2.350780966144005,
|
| 1519 |
+
"learning_rate": 4.497929545940968e-06,
|
| 1520 |
+
"loss": 0.44,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5815831987075929,
|
| 1525 |
+
"grad_norm": 2.327312584804818,
|
| 1526 |
+
"learning_rate": 4.4511882912425214e-06,
|
| 1527 |
+
"loss": 0.4366,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5842757135164244,
|
| 1532 |
+
"grad_norm": 2.294345174675599,
|
| 1533 |
+
"learning_rate": 4.404495532667525e-06,
|
| 1534 |
+
"loss": 0.4383,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5869682283252557,
|
| 1539 |
+
"grad_norm": 2.2174354559341767,
|
| 1540 |
+
"learning_rate": 4.357855396253648e-06,
|
| 1541 |
+
"loss": 0.4412,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5896607431340872,
|
| 1546 |
+
"grad_norm": 2.350816369766935,
|
| 1547 |
+
"learning_rate": 4.311272003388569e-06,
|
| 1548 |
+
"loss": 0.442,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5923532579429187,
|
| 1553 |
+
"grad_norm": 2.3824872125165286,
|
| 1554 |
+
"learning_rate": 4.2647494704457775e-06,
|
| 1555 |
+
"loss": 0.4349,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5950457727517501,
|
| 1560 |
+
"grad_norm": 2.2923371304598605,
|
| 1561 |
+
"learning_rate": 4.218291908420842e-06,
|
| 1562 |
+
"loss": 0.4286,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5977382875605816,
|
| 1567 |
+
"grad_norm": 2.346360268676029,
|
| 1568 |
+
"learning_rate": 4.171903422568128e-06,
|
| 1569 |
+
"loss": 0.4336,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.600430802369413,
|
| 1574 |
+
"grad_norm": 2.218093217821692,
|
| 1575 |
+
"learning_rate": 4.1255881120380335e-06,
|
| 1576 |
+
"loss": 0.446,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6031233171782445,
|
| 1581 |
+
"grad_norm": 2.276362681621089,
|
| 1582 |
+
"learning_rate": 4.079350069514779e-06,
|
| 1583 |
+
"loss": 0.4407,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6058158319870759,
|
| 1588 |
+
"grad_norm": 2.3056261390539934,
|
| 1589 |
+
"learning_rate": 4.033193380854733e-06,
|
| 1590 |
+
"loss": 0.4289,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6085083467959074,
|
| 1595 |
+
"grad_norm": 2.1989394458411686,
|
| 1596 |
+
"learning_rate": 3.9871221247253835e-06,
|
| 1597 |
+
"loss": 0.4175,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6112008616047389,
|
| 1602 |
+
"grad_norm": 2.3562705886952915,
|
| 1603 |
+
"learning_rate": 3.9411403722449075e-06,
|
| 1604 |
+
"loss": 0.4259,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6138933764135702,
|
| 1609 |
+
"grad_norm": 2.322065883191611,
|
| 1610 |
+
"learning_rate": 3.895252186622433e-06,
|
| 1611 |
+
"loss": 0.418,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6165858912224017,
|
| 1616 |
+
"grad_norm": 2.315390737203203,
|
| 1617 |
+
"learning_rate": 3.849461622798993e-06,
|
| 1618 |
+
"loss": 0.4174,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6192784060312332,
|
| 1623 |
+
"grad_norm": 2.136385235781553,
|
| 1624 |
+
"learning_rate": 3.8037727270891927e-06,
|
| 1625 |
+
"loss": 0.4182,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6219709208400647,
|
| 1630 |
+
"grad_norm": 2.2676350954914257,
|
| 1631 |
+
"learning_rate": 3.758189536823673e-06,
|
| 1632 |
+
"loss": 0.4162,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.624663435648896,
|
| 1637 |
+
"grad_norm": 2.3081143159586164,
|
| 1638 |
+
"learning_rate": 3.71271607999234e-06,
|
| 1639 |
+
"loss": 0.415,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6273559504577275,
|
| 1644 |
+
"grad_norm": 2.1196383763132784,
|
| 1645 |
+
"learning_rate": 3.6673563748884268e-06,
|
| 1646 |
+
"loss": 0.4225,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.630048465266559,
|
| 1651 |
+
"grad_norm": 2.3577148406371284,
|
| 1652 |
+
"learning_rate": 3.6221144297534178e-06,
|
| 1653 |
+
"loss": 0.41,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6327409800753904,
|
| 1658 |
+
"grad_norm": 2.332635685786891,
|
| 1659 |
+
"learning_rate": 3.5769942424228578e-06,
|
| 1660 |
+
"loss": 0.4137,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6354334948842219,
|
| 1665 |
+
"grad_norm": 2.254772419320123,
|
| 1666 |
+
"learning_rate": 3.531999799973079e-06,
|
| 1667 |
+
"loss": 0.429,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6381260096930533,
|
| 1672 |
+
"grad_norm": 2.1991733799648907,
|
| 1673 |
+
"learning_rate": 3.4871350783688795e-06,
|
| 1674 |
+
"loss": 0.3971,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6408185245018848,
|
| 1679 |
+
"grad_norm": 2.2671939485505916,
|
| 1680 |
+
"learning_rate": 3.4424040421121903e-06,
|
| 1681 |
+
"loss": 0.4163,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6435110393107162,
|
| 1686 |
+
"grad_norm": 2.258087169479648,
|
| 1687 |
+
"learning_rate": 3.397810643891741e-06,
|
| 1688 |
+
"loss": 0.4059,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6462035541195477,
|
| 1693 |
+
"grad_norm": 2.3543023457270325,
|
| 1694 |
+
"learning_rate": 3.35335882423378e-06,
|
| 1695 |
+
"loss": 0.4224,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6488960689283791,
|
| 1700 |
+
"grad_norm": 2.146955492636297,
|
| 1701 |
+
"learning_rate": 3.309052511153873e-06,
|
| 1702 |
+
"loss": 0.4087,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6515885837372105,
|
| 1707 |
+
"grad_norm": 2.238937583269941,
|
| 1708 |
+
"learning_rate": 3.2648956198097946e-06,
|
| 1709 |
+
"loss": 0.4053,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.654281098546042,
|
| 1714 |
+
"grad_norm": 2.190819239320771,
|
| 1715 |
+
"learning_rate": 3.2208920521555677e-06,
|
| 1716 |
+
"loss": 0.3945,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6569736133548735,
|
| 1721 |
+
"grad_norm": 2.1827434556639442,
|
| 1722 |
+
"learning_rate": 3.177045696596658e-06,
|
| 1723 |
+
"loss": 0.4089,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6596661281637048,
|
| 1728 |
+
"grad_norm": 2.128197496062037,
|
| 1729 |
+
"learning_rate": 3.133360427646378e-06,
|
| 1730 |
+
"loss": 0.3943,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6623586429725363,
|
| 1735 |
+
"grad_norm": 2.258181737280764,
|
| 1736 |
+
"learning_rate": 3.089840105583516e-06,
|
| 1737 |
+
"loss": 0.4014,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6650511577813678,
|
| 1742 |
+
"grad_norm": 2.2263287080503202,
|
| 1743 |
+
"learning_rate": 3.0464885761112083e-06,
|
| 1744 |
+
"loss": 0.4085,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6677436725901993,
|
| 1749 |
+
"grad_norm": 2.1158495070811467,
|
| 1750 |
+
"learning_rate": 3.0033096700171218e-06,
|
| 1751 |
+
"loss": 0.3919,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6704361873990307,
|
| 1756 |
+
"grad_norm": 2.1418686648492415,
|
| 1757 |
+
"learning_rate": 2.960307202834941e-06,
|
| 1758 |
+
"loss": 0.408,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6731287022078621,
|
| 1763 |
+
"grad_norm": 2.13102582996017,
|
| 1764 |
+
"learning_rate": 2.9174849745072003e-06,
|
| 1765 |
+
"loss": 0.3941,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6758212170166936,
|
| 1770 |
+
"grad_norm": 2.1255705801818277,
|
| 1771 |
+
"learning_rate": 2.874846769049509e-06,
|
| 1772 |
+
"loss": 0.4043,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.678513731825525,
|
| 1777 |
+
"grad_norm": 2.314046111156198,
|
| 1778 |
+
"learning_rate": 2.8323963542161665e-06,
|
| 1779 |
+
"loss": 0.4049,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6812062466343565,
|
| 1784 |
+
"grad_norm": 2.247094999606519,
|
| 1785 |
+
"learning_rate": 2.7901374811672256e-06,
|
| 1786 |
+
"loss": 0.3924,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.683898761443188,
|
| 1791 |
+
"grad_norm": 2.1331620184210958,
|
| 1792 |
+
"learning_rate": 2.748073884137016e-06,
|
| 1793 |
+
"loss": 0.3942,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6865912762520194,
|
| 1798 |
+
"grad_norm": 2.2729136222517257,
|
| 1799 |
+
"learning_rate": 2.7062092801041717e-06,
|
| 1800 |
+
"loss": 0.39,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6892837910608508,
|
| 1805 |
+
"grad_norm": 2.1267931341347968,
|
| 1806 |
+
"learning_rate": 2.664547368463171e-06,
|
| 1807 |
+
"loss": 0.3833,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6919763058696823,
|
| 1812 |
+
"grad_norm": 2.1728485155426838,
|
| 1813 |
+
"learning_rate": 2.6230918306974378e-06,
|
| 1814 |
+
"loss": 0.3929,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6946688206785138,
|
| 1819 |
+
"grad_norm": 2.247179993428309,
|
| 1820 |
+
"learning_rate": 2.581846330054034e-06,
|
| 1821 |
+
"loss": 0.3915,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6973613354873451,
|
| 1826 |
+
"grad_norm": 2.146951309924855,
|
| 1827 |
+
"learning_rate": 2.5408145112199424e-06,
|
| 1828 |
+
"loss": 0.3894,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7000538502961766,
|
| 1833 |
+
"grad_norm": 2.1704461453183628,
|
| 1834 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1835 |
+
"loss": 0.3942,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7027463651050081,
|
| 1840 |
+
"grad_norm": 2.164674877249693,
|
| 1841 |
+
"learning_rate": 2.4594064029965197e-06,
|
| 1842 |
+
"loss": 0.3883,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7054388799138396,
|
| 1847 |
+
"grad_norm": 2.2000767508999175,
|
| 1848 |
+
"learning_rate": 2.4190373072905687e-06,
|
| 1849 |
+
"loss": 0.3884,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.708131394722671,
|
| 1854 |
+
"grad_norm": 2.3684557762781546,
|
| 1855 |
+
"learning_rate": 2.378896280125006e-06,
|
| 1856 |
+
"loss": 0.3864,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7108239095315024,
|
| 1861 |
+
"grad_norm": 1.98524822133492,
|
| 1862 |
+
"learning_rate": 2.3389868685892573e-06,
|
| 1863 |
+
"loss": 0.3704,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7135164243403339,
|
| 1868 |
+
"grad_norm": 2.19288217749482,
|
| 1869 |
+
"learning_rate": 2.2993125993058727e-06,
|
| 1870 |
+
"loss": 0.3747,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7162089391491653,
|
| 1875 |
+
"grad_norm": 2.1617200756921497,
|
| 1876 |
+
"learning_rate": 2.2598769781189004e-06,
|
| 1877 |
+
"loss": 0.3865,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7189014539579968,
|
| 1882 |
+
"grad_norm": 2.132621198575595,
|
| 1883 |
+
"learning_rate": 2.2206834897840814e-06,
|
| 1884 |
+
"loss": 0.3871,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7215939687668282,
|
| 1889 |
+
"grad_norm": 2.145671425646657,
|
| 1890 |
+
"learning_rate": 2.1817355976609212e-06,
|
| 1891 |
+
"loss": 0.3865,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7242864835756596,
|
| 1896 |
+
"grad_norm": 2.0530895422503868,
|
| 1897 |
+
"learning_rate": 2.14303674340665e-06,
|
| 1898 |
+
"loss": 0.3673,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7269789983844911,
|
| 1903 |
+
"grad_norm": 2.2838748871806054,
|
| 1904 |
+
"learning_rate": 2.1045903466720915e-06,
|
| 1905 |
+
"loss": 0.38,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7296715131933226,
|
| 1910 |
+
"grad_norm": 2.2502198899358348,
|
| 1911 |
+
"learning_rate": 2.0663998047994906e-06,
|
| 1912 |
+
"loss": 0.3874,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.732364028002154,
|
| 1917 |
+
"grad_norm": 2.128622996026954,
|
| 1918 |
+
"learning_rate": 2.0284684925223006e-06,
|
| 1919 |
+
"loss": 0.3775,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7350565428109854,
|
| 1924 |
+
"grad_norm": 2.219696821235303,
|
| 1925 |
+
"learning_rate": 1.990799761666975e-06,
|
| 1926 |
+
"loss": 0.3798,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7377490576198169,
|
| 1931 |
+
"grad_norm": 2.197304231151606,
|
| 1932 |
+
"learning_rate": 1.953396940856776e-06,
|
| 1933 |
+
"loss": 0.3724,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7404415724286484,
|
| 1938 |
+
"grad_norm": 2.2465171659133674,
|
| 1939 |
+
"learning_rate": 1.9162633352176484e-06,
|
| 1940 |
+
"loss": 0.3831,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7431340872374798,
|
| 1945 |
+
"grad_norm": 2.1996595546449367,
|
| 1946 |
+
"learning_rate": 1.8794022260861483e-06,
|
| 1947 |
+
"loss": 0.383,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7458266020463112,
|
| 1952 |
+
"grad_norm": 2.1017548100685417,
|
| 1953 |
+
"learning_rate": 1.8428168707194922e-06,
|
| 1954 |
+
"loss": 0.3765,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7485191168551427,
|
| 1959 |
+
"grad_norm": 2.064327514284528,
|
| 1960 |
+
"learning_rate": 1.8065105020077262e-06,
|
| 1961 |
+
"loss": 0.3694,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7512116316639742,
|
| 1966 |
+
"grad_norm": 2.0962178314477016,
|
| 1967 |
+
"learning_rate": 1.7704863281880496e-06,
|
| 1968 |
+
"loss": 0.3866,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7539041464728056,
|
| 1973 |
+
"grad_norm": 2.124206708166897,
|
| 1974 |
+
"learning_rate": 1.7347475325613134e-06,
|
| 1975 |
+
"loss": 0.3741,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.756596661281637,
|
| 1980 |
+
"grad_norm": 2.379442181939827,
|
| 1981 |
+
"learning_rate": 1.6992972732107322e-06,
|
| 1982 |
+
"loss": 0.3762,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7592891760904685,
|
| 1987 |
+
"grad_norm": 2.2436516667976933,
|
| 1988 |
+
"learning_rate": 1.6641386827228107e-06,
|
| 1989 |
+
"loss": 0.3668,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7619816908992999,
|
| 1994 |
+
"grad_norm": 2.106915257521231,
|
| 1995 |
+
"learning_rate": 1.6292748679105407e-06,
|
| 1996 |
+
"loss": 0.3715,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7646742057081314,
|
| 2001 |
+
"grad_norm": 2.0865052999042546,
|
| 2002 |
+
"learning_rate": 1.594708909538854e-06,
|
| 2003 |
+
"loss": 0.3798,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7673667205169629,
|
| 2008 |
+
"grad_norm": 2.2132376460253753,
|
| 2009 |
+
"learning_rate": 1.5604438620523932e-06,
|
| 2010 |
+
"loss": 0.3641,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7700592353257943,
|
| 2015 |
+
"grad_norm": 2.101010632458687,
|
| 2016 |
+
"learning_rate": 1.5264827533056058e-06,
|
| 2017 |
+
"loss": 0.3736,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7727517501346257,
|
| 2022 |
+
"grad_norm": 2.0835331975727573,
|
| 2023 |
+
"learning_rate": 1.492828584295184e-06,
|
| 2024 |
+
"loss": 0.3569,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7754442649434572,
|
| 2029 |
+
"grad_norm": 2.265759304633531,
|
| 2030 |
+
"learning_rate": 1.4594843288948773e-06,
|
| 2031 |
+
"loss": 0.3662,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7781367797522887,
|
| 2036 |
+
"grad_norm": 2.2173463946322323,
|
| 2037 |
+
"learning_rate": 1.4264529335927114e-06,
|
| 2038 |
+
"loss": 0.37,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.78082929456112,
|
| 2043 |
+
"grad_norm": 2.1518226151098125,
|
| 2044 |
+
"learning_rate": 1.3937373172306056e-06,
|
| 2045 |
+
"loss": 0.3638,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7835218093699515,
|
| 2050 |
+
"grad_norm": 2.055682580111425,
|
| 2051 |
+
"learning_rate": 1.361340370746464e-06,
|
| 2052 |
+
"loss": 0.3702,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.786214324178783,
|
| 2057 |
+
"grad_norm": 2.0945805072566586,
|
| 2058 |
+
"learning_rate": 1.3292649569187072e-06,
|
| 2059 |
+
"loss": 0.3641,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7889068389876144,
|
| 2064 |
+
"grad_norm": 2.099143771457234,
|
| 2065 |
+
"learning_rate": 1.2975139101132999e-06,
|
| 2066 |
+
"loss": 0.3647,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7915993537964459,
|
| 2071 |
+
"grad_norm": 2.2165194526856378,
|
| 2072 |
+
"learning_rate": 1.2660900360332927e-06,
|
| 2073 |
+
"loss": 0.3715,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7942918686052773,
|
| 2078 |
+
"grad_norm": 2.244374868499319,
|
| 2079 |
+
"learning_rate": 1.2349961114708936e-06,
|
| 2080 |
+
"loss": 0.3623,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7969843834141088,
|
| 2085 |
+
"grad_norm": 2.186870000073866,
|
| 2086 |
+
"learning_rate": 1.2042348840620955e-06,
|
| 2087 |
+
"loss": 0.3729,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7996768982229402,
|
| 2092 |
+
"grad_norm": 2.5050253183494244,
|
| 2093 |
+
"learning_rate": 1.1738090720438782e-06,
|
| 2094 |
+
"loss": 0.3576,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8023694130317717,
|
| 2099 |
+
"grad_norm": 2.0739438241919594,
|
| 2100 |
+
"learning_rate": 1.1437213640140105e-06,
|
| 2101 |
+
"loss": 0.3617,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8050619278406032,
|
| 2106 |
+
"grad_norm": 2.0754012982092265,
|
| 2107 |
+
"learning_rate": 1.113974418693471e-06,
|
| 2108 |
+
"loss": 0.3595,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8077544426494345,
|
| 2113 |
+
"grad_norm": 2.3081364092115133,
|
| 2114 |
+
"learning_rate": 1.0845708646915054e-06,
|
| 2115 |
+
"loss": 0.361,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.810446957458266,
|
| 2120 |
+
"grad_norm": 2.0032280348669067,
|
| 2121 |
+
"learning_rate": 1.055513300273352e-06,
|
| 2122 |
+
"loss": 0.3578,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8131394722670975,
|
| 2127 |
+
"grad_norm": 2.0937736954570365,
|
| 2128 |
+
"learning_rate": 1.026804293130641e-06,
|
| 2129 |
+
"loss": 0.3616,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.815831987075929,
|
| 2134 |
+
"grad_norm": 2.1070535051869443,
|
| 2135 |
+
"learning_rate": 9.984463801544992e-07,
|
| 2136 |
+
"loss": 0.3535,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8185245018847603,
|
| 2141 |
+
"grad_norm": 2.3516801013906097,
|
| 2142 |
+
"learning_rate": 9.704420672113774e-07,
|
| 2143 |
+
"loss": 0.3536,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8212170166935918,
|
| 2148 |
+
"grad_norm": 1.9740143572115736,
|
| 2149 |
+
"learning_rate": 9.42793828921616e-07,
|
| 2150 |
+
"loss": 0.3608,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8239095315024233,
|
| 2155 |
+
"grad_norm": 2.0979215240658027,
|
| 2156 |
+
"learning_rate": 9.15504108440774e-07,
|
| 2157 |
+
"loss": 0.3736,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8266020463112547,
|
| 2162 |
+
"grad_norm": 2.0284760942560323,
|
| 2163 |
+
"learning_rate": 8.885753172437389e-07,
|
| 2164 |
+
"loss": 0.3508,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8292945611200861,
|
| 2169 |
+
"grad_norm": 2.0211247506598395,
|
| 2170 |
+
"learning_rate": 8.620098349116329e-07,
|
| 2171 |
+
"loss": 0.352,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8319870759289176,
|
| 2176 |
+
"grad_norm": 2.3428781214119985,
|
| 2177 |
+
"learning_rate": 8.358100089215426e-07,
|
| 2178 |
+
"loss": 0.3603,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8346795907377491,
|
| 2183 |
+
"grad_norm": 2.109207498019035,
|
| 2184 |
+
"learning_rate": 8.099781544390795e-07,
|
| 2185 |
+
"loss": 0.3629,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8373721055465805,
|
| 2190 |
+
"grad_norm": 2.100615618738898,
|
| 2191 |
+
"learning_rate": 7.845165541138011e-07,
|
| 2192 |
+
"loss": 0.3556,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.840064620355412,
|
| 2197 |
+
"grad_norm": 2.0303755145271243,
|
| 2198 |
+
"learning_rate": 7.594274578775007e-07,
|
| 2199 |
+
"loss": 0.3555,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8427571351642434,
|
| 2204 |
+
"grad_norm": 2.0652917821688104,
|
| 2205 |
+
"learning_rate": 7.347130827453941e-07,
|
| 2206 |
+
"loss": 0.3588,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8454496499730748,
|
| 2211 |
+
"grad_norm": 1.9882512618837487,
|
| 2212 |
+
"learning_rate": 7.103756126202071e-07,
|
| 2213 |
+
"loss": 0.3613,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8481421647819063,
|
| 2218 |
+
"grad_norm": 2.067179222751156,
|
| 2219 |
+
"learning_rate": 6.864171980991985e-07,
|
| 2220 |
+
"loss": 0.3453,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8508346795907378,
|
| 2225 |
+
"grad_norm": 2.093050443455704,
|
| 2226 |
+
"learning_rate": 6.628399562841159e-07,
|
| 2227 |
+
"loss": 0.3548,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8535271943995691,
|
| 2232 |
+
"grad_norm": 2.1033534008065073,
|
| 2233 |
+
"learning_rate": 6.396459705941204e-07,
|
| 2234 |
+
"loss": 0.3471,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8562197092084006,
|
| 2239 |
+
"grad_norm": 1.9167623533757014,
|
| 2240 |
+
"learning_rate": 6.168372905816822e-07,
|
| 2241 |
+
"loss": 0.3477,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8589122240172321,
|
| 2246 |
+
"grad_norm": 2.1430265742267105,
|
| 2247 |
+
"learning_rate": 5.944159317514709e-07,
|
| 2248 |
+
"loss": 0.3517,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8616047388260636,
|
| 2253 |
+
"grad_norm": 2.1988162518545686,
|
| 2254 |
+
"learning_rate": 5.723838753822525e-07,
|
| 2255 |
+
"loss": 0.3689,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.864297253634895,
|
| 2260 |
+
"grad_norm": 2.2356747531202807,
|
| 2261 |
+
"learning_rate": 5.507430683518161e-07,
|
| 2262 |
+
"loss": 0.3449,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8669897684437264,
|
| 2267 |
+
"grad_norm": 2.1308682594997324,
|
| 2268 |
+
"learning_rate": 5.294954229649329e-07,
|
| 2269 |
+
"loss": 0.3507,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8696822832525579,
|
| 2274 |
+
"grad_norm": 2.0875409392190107,
|
| 2275 |
+
"learning_rate": 5.086428167843755e-07,
|
| 2276 |
+
"loss": 0.3476,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8723747980613893,
|
| 2281 |
+
"grad_norm": 2.147966222762087,
|
| 2282 |
+
"learning_rate": 4.881870924650062e-07,
|
| 2283 |
+
"loss": 0.342,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8750673128702208,
|
| 2288 |
+
"grad_norm": 2.0556458342996464,
|
| 2289 |
+
"learning_rate": 4.681300575909492e-07,
|
| 2290 |
+
"loss": 0.3471,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8777598276790523,
|
| 2295 |
+
"grad_norm": 2.015550654050713,
|
| 2296 |
+
"learning_rate": 4.4847348451586226e-07,
|
| 2297 |
+
"loss": 0.3478,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8804523424878837,
|
| 2302 |
+
"grad_norm": 1.998136493936924,
|
| 2303 |
+
"learning_rate": 4.2921911020631926e-07,
|
| 2304 |
+
"loss": 0.3503,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8831448572967151,
|
| 2309 |
+
"grad_norm": 1.8537901854708771,
|
| 2310 |
+
"learning_rate": 4.1036863608832856e-07,
|
| 2311 |
+
"loss": 0.3468,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8858373721055466,
|
| 2316 |
+
"grad_norm": 2.080626726929853,
|
| 2317 |
+
"learning_rate": 3.919237278969773e-07,
|
| 2318 |
+
"loss": 0.3445,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8885298869143781,
|
| 2323 |
+
"grad_norm": 1.8500187551625011,
|
| 2324 |
+
"learning_rate": 3.7388601552924066e-07,
|
| 2325 |
+
"loss": 0.3515,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8912224017232094,
|
| 2330 |
+
"grad_norm": 1.9930943811147626,
|
| 2331 |
+
"learning_rate": 3.56257092899957e-07,
|
| 2332 |
+
"loss": 0.3389,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8939149165320409,
|
| 2337 |
+
"grad_norm": 2.053675256803709,
|
| 2338 |
+
"learning_rate": 3.390385178009781e-07,
|
| 2339 |
+
"loss": 0.3534,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8966074313408724,
|
| 2344 |
+
"grad_norm": 2.0176018019009225,
|
| 2345 |
+
"learning_rate": 3.222318117635143e-07,
|
| 2346 |
+
"loss": 0.3391,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8992999461497039,
|
| 2351 |
+
"grad_norm": 2.165084875616374,
|
| 2352 |
+
"learning_rate": 3.058384599236847e-07,
|
| 2353 |
+
"loss": 0.3392,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9019924609585352,
|
| 2358 |
+
"grad_norm": 1.8300243137706005,
|
| 2359 |
+
"learning_rate": 2.8985991089127916e-07,
|
| 2360 |
+
"loss": 0.3519,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9046849757673667,
|
| 2365 |
+
"grad_norm": 1.9039488415439678,
|
| 2366 |
+
"learning_rate": 2.7429757662175316e-07,
|
| 2367 |
+
"loss": 0.3362,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9073774905761982,
|
| 2372 |
+
"grad_norm": 2.1498883952955805,
|
| 2373 |
+
"learning_rate": 2.5915283229146106e-07,
|
| 2374 |
+
"loss": 0.3485,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9100700053850296,
|
| 2379 |
+
"grad_norm": 1.9985654494313574,
|
| 2380 |
+
"learning_rate": 2.44427016176132e-07,
|
| 2381 |
+
"loss": 0.349,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9127625201938611,
|
| 2386 |
+
"grad_norm": 2.2813576484209896,
|
| 2387 |
+
"learning_rate": 2.301214295326193e-07,
|
| 2388 |
+
"loss": 0.3462,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9154550350026925,
|
| 2393 |
+
"grad_norm": 2.0155307395335385,
|
| 2394 |
+
"learning_rate": 2.162373364839071e-07,
|
| 2395 |
+
"loss": 0.334,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9181475498115239,
|
| 2400 |
+
"grad_norm": 1.865462382587576,
|
| 2401 |
+
"learning_rate": 2.0277596390741005e-07,
|
| 2402 |
+
"loss": 0.3465,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9208400646203554,
|
| 2407 |
+
"grad_norm": 1.9778403222139225,
|
| 2408 |
+
"learning_rate": 1.8973850132655957e-07,
|
| 2409 |
+
"loss": 0.3399,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9235325794291869,
|
| 2414 |
+
"grad_norm": 2.1982054104516986,
|
| 2415 |
+
"learning_rate": 1.7712610080568738e-07,
|
| 2416 |
+
"loss": 0.3471,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9262250942380184,
|
| 2421 |
+
"grad_norm": 2.0362535026984396,
|
| 2422 |
+
"learning_rate": 1.649398768482241e-07,
|
| 2423 |
+
"loss": 0.35,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9289176090468497,
|
| 2428 |
+
"grad_norm": 2.0586480283901714,
|
| 2429 |
+
"learning_rate": 1.5318090629821757e-07,
|
| 2430 |
+
"loss": 0.3414,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9316101238556812,
|
| 2435 |
+
"grad_norm": 2.0363932956107,
|
| 2436 |
+
"learning_rate": 1.4185022824517392e-07,
|
| 2437 |
+
"loss": 0.3432,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9343026386645127,
|
| 2442 |
+
"grad_norm": 2.0787927070430943,
|
| 2443 |
+
"learning_rate": 1.3094884393223872e-07,
|
| 2444 |
+
"loss": 0.3491,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9369951534733441,
|
| 2449 |
+
"grad_norm": 1.9680040636502156,
|
| 2450 |
+
"learning_rate": 1.2047771666772124e-07,
|
| 2451 |
+
"loss": 0.3452,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9396876682821755,
|
| 2456 |
+
"grad_norm": 1.8963738671804522,
|
| 2457 |
+
"learning_rate": 1.1043777173997239e-07,
|
| 2458 |
+
"loss": 0.3437,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.942380183091007,
|
| 2463 |
+
"grad_norm": 2.074849031085683,
|
| 2464 |
+
"learning_rate": 1.0082989633561746e-07,
|
| 2465 |
+
"loss": 0.3399,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9450726978998385,
|
| 2470 |
+
"grad_norm": 2.005988363818133,
|
| 2471 |
+
"learning_rate": 9.165493946116432e-08,
|
| 2472 |
+
"loss": 0.341,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9477652127086699,
|
| 2477 |
+
"grad_norm": 1.9739191876783688,
|
| 2478 |
+
"learning_rate": 8.29137118679757e-08,
|
| 2479 |
+
"loss": 0.3505,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9504577275175013,
|
| 2484 |
+
"grad_norm": 2.0544856800909863,
|
| 2485 |
+
"learning_rate": 7.460698598063032e-08,
|
| 2486 |
+
"loss": 0.3504,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9531502423263328,
|
| 2491 |
+
"grad_norm": 2.0266695028121604,
|
| 2492 |
+
"learning_rate": 6.673549582866368e-08,
|
| 2493 |
+
"loss": 0.3437,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9558427571351642,
|
| 2498 |
+
"grad_norm": 1.9158086758881532,
|
| 2499 |
+
"learning_rate": 5.929993698170877e-08,
|
| 2500 |
+
"loss": 0.3407,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9585352719439957,
|
| 2505 |
+
"grad_norm": 2.060993161808353,
|
| 2506 |
+
"learning_rate": 5.2300966488026406e-08,
|
| 2507 |
+
"loss": 0.3514,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9612277867528272,
|
| 2512 |
+
"grad_norm": 2.2300563421538735,
|
| 2513 |
+
"learning_rate": 4.573920281645161e-08,
|
| 2514 |
+
"loss": 0.3329,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9639203015616586,
|
| 2519 |
+
"grad_norm": 1.9501346199167315,
|
| 2520 |
+
"learning_rate": 3.96152258017346e-08,
|
| 2521 |
+
"loss": 0.3373,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.96661281637049,
|
| 2526 |
+
"grad_norm": 2.078047814961423,
|
| 2527 |
+
"learning_rate": 3.3929576593310644e-08,
|
| 2528 |
+
"loss": 0.3498,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9693053311793215,
|
| 2533 |
+
"grad_norm": 1.9588663874547265,
|
| 2534 |
+
"learning_rate": 2.868275760747441e-08,
|
| 2535 |
+
"loss": 0.3351,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.971997845988153,
|
| 2540 |
+
"grad_norm": 2.0555012024055563,
|
| 2541 |
+
"learning_rate": 2.387523248298773e-08,
|
| 2542 |
+
"loss": 0.3438,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9746903607969843,
|
| 2547 |
+
"grad_norm": 2.02985190947748,
|
| 2548 |
+
"learning_rate": 1.9507426040110134e-08,
|
| 2549 |
+
"loss": 0.3371,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9773828756058158,
|
| 2554 |
+
"grad_norm": 1.9411470270835758,
|
| 2555 |
+
"learning_rate": 1.557972424305665e-08,
|
| 2556 |
+
"loss": 0.3423,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9800753904146473,
|
| 2561 |
+
"grad_norm": 2.1196737223933773,
|
| 2562 |
+
"learning_rate": 1.209247416589232e-08,
|
| 2563 |
+
"loss": 0.344,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9827679052234787,
|
| 2568 |
+
"grad_norm": 2.051817684549473,
|
| 2569 |
+
"learning_rate": 9.045983961865622e-09,
|
| 2570 |
+
"loss": 0.3568,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9854604200323102,
|
| 2575 |
+
"grad_norm": 2.0189362221244957,
|
| 2576 |
+
"learning_rate": 6.440522836174135e-09,
|
| 2577 |
+
"loss": 0.3428,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9881529348411416,
|
| 2582 |
+
"grad_norm": 1.9669363709037755,
|
| 2583 |
+
"learning_rate": 4.27632102218023e-09,
|
| 2584 |
+
"loss": 0.3344,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9908454496499731,
|
| 2589 |
+
"grad_norm": 1.905202601388718,
|
| 2590 |
+
"learning_rate": 2.5535697610618026e-09,
|
| 2591 |
+
"loss": 0.3398,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9935379644588045,
|
| 2596 |
+
"grad_norm": 1.9833936251009352,
|
| 2597 |
+
"learning_rate": 1.2724212849180019e-09,
|
| 2598 |
+
"loss": 0.3458,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.996230479267636,
|
| 2603 |
+
"grad_norm": 1.8216154780128988,
|
| 2604 |
+
"learning_rate": 4.329888033127727e-10,
|
| 2605 |
+
"loss": 0.3413,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9989229940764675,
|
| 2610 |
+
"grad_norm": 2.0143818367459474,
|
| 2611 |
+
"learning_rate": 3.5346493273968886e-11,
|
| 2612 |
+
"loss": 0.3454,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 1.0,
|
| 2617 |
+
"eval_runtime": 3.4477,
|
| 2618 |
+
"eval_samples_per_second": 2.9,
|
| 2619 |
+
"eval_steps_per_second": 0.87,
|
| 2620 |
+
"step": 1857
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 1.0,
|
| 2624 |
+
"step": 1857,
|
| 2625 |
+
"total_flos": 194409010298880.0,
|
| 2626 |
+
"train_loss": 0.5086988852623655,
|
| 2627 |
+
"train_runtime": 16625.9643,
|
| 2628 |
+
"train_samples_per_second": 1.787,
|
| 2629 |
+
"train_steps_per_second": 0.112
|
| 2630 |
+
}
|
| 2631 |
+
],
|
| 2632 |
+
"logging_steps": 5,
|
| 2633 |
+
"max_steps": 1857,
|
| 2634 |
+
"num_input_tokens_seen": 0,
|
| 2635 |
+
"num_train_epochs": 1,
|
| 2636 |
+
"save_steps": 100,
|
| 2637 |
+
"stateful_callbacks": {
|
| 2638 |
+
"TrainerControl": {
|
| 2639 |
+
"args": {
|
| 2640 |
+
"should_epoch_stop": false,
|
| 2641 |
+
"should_evaluate": false,
|
| 2642 |
+
"should_log": false,
|
| 2643 |
+
"should_save": true,
|
| 2644 |
+
"should_training_stop": true
|
| 2645 |
+
},
|
| 2646 |
+
"attributes": {}
|
| 2647 |
+
}
|
| 2648 |
+
},
|
| 2649 |
+
"total_flos": 194409010298880.0,
|
| 2650 |
+
"train_batch_size": 4,
|
| 2651 |
+
"trial_name": null,
|
| 2652 |
+
"trial_params": null
|
| 2653 |
+
}
|