Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep24_01-42-01_COE-CS-sv003/events.out.tfevents.1727142428.COE-CS-sv003.751827.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2660 -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_syntaxcleanbatch16_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_syntaxcleanbatch16_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": 195037149265920.0,
|
| 4 |
+
"train_loss": 0.5120102114659513,
|
| 5 |
+
"train_runtime": 16685.4185,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.786,
|
| 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_01-42-01_COE-CS-sv003/events.out.tfevents.1727142428.COE-CS-sv003.751827.0
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 sha256:9b9a10c8a3f7deec1fbbdb602b351f1d51bbb841fb46c0b2fcf29f1cd798c8d3
|
| 3 |
+
size 84482
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 195037149265920.0,
|
| 4 |
+
"train_loss": 0.5120102114659513,
|
| 5 |
+
"train_runtime": 16685.4185,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.786,
|
| 8 |
+
"train_steps_per_second": 0.112
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2660 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1863,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005367686527106817,
|
| 13 |
+
"grad_norm": 24.305655426812773,
|
| 14 |
+
"learning_rate": 5.3475935828877005e-08,
|
| 15 |
+
"loss": 1.3555,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0026838432635534087,
|
| 20 |
+
"grad_norm": 23.22138878044751,
|
| 21 |
+
"learning_rate": 2.6737967914438503e-07,
|
| 22 |
+
"loss": 1.3448,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005367686527106817,
|
| 27 |
+
"grad_norm": 15.39412500867573,
|
| 28 |
+
"learning_rate": 5.347593582887701e-07,
|
| 29 |
+
"loss": 1.3152,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008051529790660225,
|
| 34 |
+
"grad_norm": 11.814690159057262,
|
| 35 |
+
"learning_rate": 8.021390374331551e-07,
|
| 36 |
+
"loss": 1.1495,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010735373054213635,
|
| 41 |
+
"grad_norm": 9.400955813538365,
|
| 42 |
+
"learning_rate": 1.0695187165775401e-06,
|
| 43 |
+
"loss": 1.0489,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013419216317767043,
|
| 48 |
+
"grad_norm": 3.612620121520767,
|
| 49 |
+
"learning_rate": 1.3368983957219254e-06,
|
| 50 |
+
"loss": 0.962,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01610305958132045,
|
| 55 |
+
"grad_norm": 3.507034366901718,
|
| 56 |
+
"learning_rate": 1.6042780748663103e-06,
|
| 57 |
+
"loss": 0.9095,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.01878690284487386,
|
| 62 |
+
"grad_norm": 2.9769449976377174,
|
| 63 |
+
"learning_rate": 1.8716577540106954e-06,
|
| 64 |
+
"loss": 0.8651,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02147074610842727,
|
| 69 |
+
"grad_norm": 2.963828184373544,
|
| 70 |
+
"learning_rate": 2.1390374331550802e-06,
|
| 71 |
+
"loss": 0.8482,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.024154589371980676,
|
| 76 |
+
"grad_norm": 3.1133147768329232,
|
| 77 |
+
"learning_rate": 2.4064171122994653e-06,
|
| 78 |
+
"loss": 0.85,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.026838432635534086,
|
| 83 |
+
"grad_norm": 2.931545369425598,
|
| 84 |
+
"learning_rate": 2.673796791443851e-06,
|
| 85 |
+
"loss": 0.8309,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.029522275899087493,
|
| 90 |
+
"grad_norm": 3.0256227773094384,
|
| 91 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 92 |
+
"loss": 0.8186,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0322061191626409,
|
| 97 |
+
"grad_norm": 3.0925781868322955,
|
| 98 |
+
"learning_rate": 3.2085561497326205e-06,
|
| 99 |
+
"loss": 0.8041,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03488996242619431,
|
| 104 |
+
"grad_norm": 3.063930015490271,
|
| 105 |
+
"learning_rate": 3.4759358288770056e-06,
|
| 106 |
+
"loss": 0.806,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03757380568974772,
|
| 111 |
+
"grad_norm": 3.2043146977534662,
|
| 112 |
+
"learning_rate": 3.7433155080213907e-06,
|
| 113 |
+
"loss": 0.8075,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.040257648953301126,
|
| 118 |
+
"grad_norm": 2.92763825791418,
|
| 119 |
+
"learning_rate": 4.010695187165775e-06,
|
| 120 |
+
"loss": 0.7952,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04294149221685454,
|
| 125 |
+
"grad_norm": 3.0927861281954665,
|
| 126 |
+
"learning_rate": 4.2780748663101604e-06,
|
| 127 |
+
"loss": 0.7924,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.045625335480407946,
|
| 132 |
+
"grad_norm": 3.0187031080233817,
|
| 133 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 134 |
+
"loss": 0.7472,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04830917874396135,
|
| 139 |
+
"grad_norm": 2.974576768839193,
|
| 140 |
+
"learning_rate": 4.812834224598931e-06,
|
| 141 |
+
"loss": 0.7615,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05099302200751476,
|
| 146 |
+
"grad_norm": 2.986145546612386,
|
| 147 |
+
"learning_rate": 5.0802139037433165e-06,
|
| 148 |
+
"loss": 0.7619,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05367686527106817,
|
| 153 |
+
"grad_norm": 3.0743318362981755,
|
| 154 |
+
"learning_rate": 5.347593582887702e-06,
|
| 155 |
+
"loss": 0.7617,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05636070853462158,
|
| 160 |
+
"grad_norm": 3.2918058213261094,
|
| 161 |
+
"learning_rate": 5.614973262032086e-06,
|
| 162 |
+
"loss": 0.7675,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.059044551798174985,
|
| 167 |
+
"grad_norm": 3.2237072710532493,
|
| 168 |
+
"learning_rate": 5.882352941176471e-06,
|
| 169 |
+
"loss": 0.7331,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06172839506172839,
|
| 174 |
+
"grad_norm": 3.1899890654845606,
|
| 175 |
+
"learning_rate": 6.149732620320856e-06,
|
| 176 |
+
"loss": 0.7459,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0644122383252818,
|
| 181 |
+
"grad_norm": 3.043596246389249,
|
| 182 |
+
"learning_rate": 6.417112299465241e-06,
|
| 183 |
+
"loss": 0.744,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06709608158883522,
|
| 188 |
+
"grad_norm": 2.9018863887175743,
|
| 189 |
+
"learning_rate": 6.684491978609626e-06,
|
| 190 |
+
"loss": 0.7377,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06977992485238863,
|
| 195 |
+
"grad_norm": 2.907463722040111,
|
| 196 |
+
"learning_rate": 6.951871657754011e-06,
|
| 197 |
+
"loss": 0.7427,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07246376811594203,
|
| 202 |
+
"grad_norm": 2.84315815544186,
|
| 203 |
+
"learning_rate": 7.219251336898396e-06,
|
| 204 |
+
"loss": 0.7319,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07514761137949544,
|
| 209 |
+
"grad_norm": 2.973654603626244,
|
| 210 |
+
"learning_rate": 7.486631016042781e-06,
|
| 211 |
+
"loss": 0.7328,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07783145464304884,
|
| 216 |
+
"grad_norm": 3.141459628944732,
|
| 217 |
+
"learning_rate": 7.754010695187166e-06,
|
| 218 |
+
"loss": 0.7338,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08051529790660225,
|
| 223 |
+
"grad_norm": 2.842095254666257,
|
| 224 |
+
"learning_rate": 8.02139037433155e-06,
|
| 225 |
+
"loss": 0.7343,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08319914117015566,
|
| 230 |
+
"grad_norm": 2.868583490076937,
|
| 231 |
+
"learning_rate": 8.288770053475937e-06,
|
| 232 |
+
"loss": 0.7287,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08588298443370908,
|
| 237 |
+
"grad_norm": 2.8470718739082255,
|
| 238 |
+
"learning_rate": 8.556149732620321e-06,
|
| 239 |
+
"loss": 0.7195,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08856682769726248,
|
| 244 |
+
"grad_norm": 2.9010358259004065,
|
| 245 |
+
"learning_rate": 8.823529411764707e-06,
|
| 246 |
+
"loss": 0.7036,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09125067096081589,
|
| 251 |
+
"grad_norm": 3.305646484034517,
|
| 252 |
+
"learning_rate": 9.090909090909091e-06,
|
| 253 |
+
"loss": 0.7215,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0939345142243693,
|
| 258 |
+
"grad_norm": 2.969331866099511,
|
| 259 |
+
"learning_rate": 9.358288770053477e-06,
|
| 260 |
+
"loss": 0.721,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.0966183574879227,
|
| 265 |
+
"grad_norm": 2.93754880819952,
|
| 266 |
+
"learning_rate": 9.625668449197861e-06,
|
| 267 |
+
"loss": 0.7038,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09930220075147611,
|
| 272 |
+
"grad_norm": 2.7050202580005065,
|
| 273 |
+
"learning_rate": 9.893048128342247e-06,
|
| 274 |
+
"loss": 0.7206,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10198604401502952,
|
| 279 |
+
"grad_norm": 3.047634989404141,
|
| 280 |
+
"learning_rate": 9.999920944317846e-06,
|
| 281 |
+
"loss": 0.6872,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10466988727858294,
|
| 286 |
+
"grad_norm": 2.795885504909657,
|
| 287 |
+
"learning_rate": 9.99943783531341e-06,
|
| 288 |
+
"loss": 0.7127,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10735373054213634,
|
| 293 |
+
"grad_norm": 2.744750472731151,
|
| 294 |
+
"learning_rate": 9.998515579512446e-06,
|
| 295 |
+
"loss": 0.7165,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.11003757380568975,
|
| 300 |
+
"grad_norm": 2.7651055012729104,
|
| 301 |
+
"learning_rate": 9.997154257925199e-06,
|
| 302 |
+
"loss": 0.7171,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11272141706924316,
|
| 307 |
+
"grad_norm": 3.206532585503099,
|
| 308 |
+
"learning_rate": 9.995353990129115e-06,
|
| 309 |
+
"loss": 0.695,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11540526033279656,
|
| 314 |
+
"grad_norm": 3.024589614754047,
|
| 315 |
+
"learning_rate": 9.99311493425834e-06,
|
| 316 |
+
"loss": 0.7037,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11808910359634997,
|
| 321 |
+
"grad_norm": 2.551935026870453,
|
| 322 |
+
"learning_rate": 9.99043728698983e-06,
|
| 323 |
+
"loss": 0.6877,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.12077294685990338,
|
| 328 |
+
"grad_norm": 2.539435880635434,
|
| 329 |
+
"learning_rate": 9.987321283526072e-06,
|
| 330 |
+
"loss": 0.7087,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12345679012345678,
|
| 335 |
+
"grad_norm": 2.676637809439421,
|
| 336 |
+
"learning_rate": 9.983767197574432e-06,
|
| 337 |
+
"loss": 0.6936,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1261406333870102,
|
| 342 |
+
"grad_norm": 2.7492296664847045,
|
| 343 |
+
"learning_rate": 9.979775341323097e-06,
|
| 344 |
+
"loss": 0.6915,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1288244766505636,
|
| 349 |
+
"grad_norm": 2.6934789882553742,
|
| 350 |
+
"learning_rate": 9.975346065413673e-06,
|
| 351 |
+
"loss": 0.7031,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.13150831991411702,
|
| 356 |
+
"grad_norm": 2.640702210698053,
|
| 357 |
+
"learning_rate": 9.970479758910365e-06,
|
| 358 |
+
"loss": 0.7108,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13419216317767044,
|
| 363 |
+
"grad_norm": 2.748234011053248,
|
| 364 |
+
"learning_rate": 9.965176849265814e-06,
|
| 365 |
+
"loss": 0.7027,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13687600644122383,
|
| 370 |
+
"grad_norm": 2.5844501833121294,
|
| 371 |
+
"learning_rate": 9.959437802283552e-06,
|
| 372 |
+
"loss": 0.6878,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13955984970477725,
|
| 377 |
+
"grad_norm": 2.66470810089039,
|
| 378 |
+
"learning_rate": 9.953263122077077e-06,
|
| 379 |
+
"loss": 0.6805,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14224369296833064,
|
| 384 |
+
"grad_norm": 2.4613007519929866,
|
| 385 |
+
"learning_rate": 9.946653351025575e-06,
|
| 386 |
+
"loss": 0.6808,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14492753623188406,
|
| 391 |
+
"grad_norm": 2.7303364431033583,
|
| 392 |
+
"learning_rate": 9.939609069726279e-06,
|
| 393 |
+
"loss": 0.6872,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14761137949543746,
|
| 398 |
+
"grad_norm": 2.4095297951330865,
|
| 399 |
+
"learning_rate": 9.932130896943477e-06,
|
| 400 |
+
"loss": 0.6788,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.15029522275899088,
|
| 405 |
+
"grad_norm": 2.5247027802281847,
|
| 406 |
+
"learning_rate": 9.924219489554145e-06,
|
| 407 |
+
"loss": 0.6903,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1529790660225443,
|
| 412 |
+
"grad_norm": 2.48852250288296,
|
| 413 |
+
"learning_rate": 9.915875542490257e-06,
|
| 414 |
+
"loss": 0.6873,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1556629092860977,
|
| 419 |
+
"grad_norm": 2.5324167659849333,
|
| 420 |
+
"learning_rate": 9.907099788677745e-06,
|
| 421 |
+
"loss": 0.6576,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.1583467525496511,
|
| 426 |
+
"grad_norm": 2.564293700320395,
|
| 427 |
+
"learning_rate": 9.897892998972113e-06,
|
| 428 |
+
"loss": 0.6685,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1610305958132045,
|
| 433 |
+
"grad_norm": 2.7292649090364183,
|
| 434 |
+
"learning_rate": 9.888255982090728e-06,
|
| 435 |
+
"loss": 0.6812,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.16371443907675792,
|
| 440 |
+
"grad_norm": 2.43813150224106,
|
| 441 |
+
"learning_rate": 9.878189584541783e-06,
|
| 442 |
+
"loss": 0.6707,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16639828234031132,
|
| 447 |
+
"grad_norm": 4.145173454039506,
|
| 448 |
+
"learning_rate": 9.867694690549943e-06,
|
| 449 |
+
"loss": 0.6579,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16908212560386474,
|
| 454 |
+
"grad_norm": 2.5276238351648552,
|
| 455 |
+
"learning_rate": 9.85677222197867e-06,
|
| 456 |
+
"loss": 0.6707,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.17176596886741816,
|
| 461 |
+
"grad_norm": 2.482449160388171,
|
| 462 |
+
"learning_rate": 9.845423138249254e-06,
|
| 463 |
+
"loss": 0.6629,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17444981213097155,
|
| 468 |
+
"grad_norm": 2.426718943488087,
|
| 469 |
+
"learning_rate": 9.833648436256525e-06,
|
| 470 |
+
"loss": 0.6677,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17713365539452497,
|
| 475 |
+
"grad_norm": 2.453880759981358,
|
| 476 |
+
"learning_rate": 9.821449150281308e-06,
|
| 477 |
+
"loss": 0.6725,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17981749865807836,
|
| 482 |
+
"grad_norm": 2.539784755342757,
|
| 483 |
+
"learning_rate": 9.808826351899551e-06,
|
| 484 |
+
"loss": 0.6626,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.18250134192163178,
|
| 489 |
+
"grad_norm": 2.519434594701762,
|
| 490 |
+
"learning_rate": 9.795781149888216e-06,
|
| 491 |
+
"loss": 0.666,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18518518518518517,
|
| 496 |
+
"grad_norm": 2.7383878655688814,
|
| 497 |
+
"learning_rate": 9.782314690127867e-06,
|
| 498 |
+
"loss": 0.6448,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1878690284487386,
|
| 503 |
+
"grad_norm": 2.64399762195394,
|
| 504 |
+
"learning_rate": 9.768428155502038e-06,
|
| 505 |
+
"loss": 0.654,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.19055287171229202,
|
| 510 |
+
"grad_norm": 2.4804265805545134,
|
| 511 |
+
"learning_rate": 9.754122765793306e-06,
|
| 512 |
+
"loss": 0.629,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.1932367149758454,
|
| 517 |
+
"grad_norm": 2.543201466486224,
|
| 518 |
+
"learning_rate": 9.739399777576169e-06,
|
| 519 |
+
"loss": 0.6409,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19592055823939883,
|
| 524 |
+
"grad_norm": 2.5609743466777397,
|
| 525 |
+
"learning_rate": 9.72426048410665e-06,
|
| 526 |
+
"loss": 0.6568,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19860440150295222,
|
| 531 |
+
"grad_norm": 2.403379282200782,
|
| 532 |
+
"learning_rate": 9.70870621520871e-06,
|
| 533 |
+
"loss": 0.645,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.20128824476650564,
|
| 538 |
+
"grad_norm": 2.3259878238624014,
|
| 539 |
+
"learning_rate": 9.692738337157441e-06,
|
| 540 |
+
"loss": 0.6507,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.20397208803005903,
|
| 545 |
+
"grad_norm": 2.5652621488474936,
|
| 546 |
+
"learning_rate": 9.676358252559034e-06,
|
| 547 |
+
"loss": 0.6441,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20665593129361245,
|
| 552 |
+
"grad_norm": 2.7173664219726934,
|
| 553 |
+
"learning_rate": 9.6595674002276e-06,
|
| 554 |
+
"loss": 0.6407,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20933977455716588,
|
| 559 |
+
"grad_norm": 2.6039911875109563,
|
| 560 |
+
"learning_rate": 9.642367255058767e-06,
|
| 561 |
+
"loss": 0.6218,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.21202361782071927,
|
| 566 |
+
"grad_norm": 2.344579879458719,
|
| 567 |
+
"learning_rate": 9.624759327900131e-06,
|
| 568 |
+
"loss": 0.6536,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2147074610842727,
|
| 573 |
+
"grad_norm": 2.6297589664652947,
|
| 574 |
+
"learning_rate": 9.606745165418554e-06,
|
| 575 |
+
"loss": 0.6363,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.21739130434782608,
|
| 580 |
+
"grad_norm": 2.39785560505727,
|
| 581 |
+
"learning_rate": 9.58832634996429e-06,
|
| 582 |
+
"loss": 0.6436,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2200751476113795,
|
| 587 |
+
"grad_norm": 2.585642851240206,
|
| 588 |
+
"learning_rate": 9.569504499432005e-06,
|
| 589 |
+
"loss": 0.6258,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2227589908749329,
|
| 594 |
+
"grad_norm": 2.8691418365076053,
|
| 595 |
+
"learning_rate": 9.550281267118659e-06,
|
| 596 |
+
"loss": 0.6275,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22544283413848631,
|
| 601 |
+
"grad_norm": 2.476684695004809,
|
| 602 |
+
"learning_rate": 9.530658341578276e-06,
|
| 603 |
+
"loss": 0.6417,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2281266774020397,
|
| 608 |
+
"grad_norm": 2.363278675306579,
|
| 609 |
+
"learning_rate": 9.510637446473633e-06,
|
| 610 |
+
"loss": 0.6242,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.23081052066559313,
|
| 615 |
+
"grad_norm": 2.4597379363178624,
|
| 616 |
+
"learning_rate": 9.490220340424844e-06,
|
| 617 |
+
"loss": 0.6375,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.23349436392914655,
|
| 622 |
+
"grad_norm": 2.383668683197536,
|
| 623 |
+
"learning_rate": 9.469408816854898e-06,
|
| 624 |
+
"loss": 0.6058,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23617820719269994,
|
| 629 |
+
"grad_norm": 2.355853732674972,
|
| 630 |
+
"learning_rate": 9.448204703832102e-06,
|
| 631 |
+
"loss": 0.6156,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23886205045625336,
|
| 636 |
+
"grad_norm": 2.3375352653671646,
|
| 637 |
+
"learning_rate": 9.426609863909537e-06,
|
| 638 |
+
"loss": 0.6238,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.24154589371980675,
|
| 643 |
+
"grad_norm": 2.547634943087699,
|
| 644 |
+
"learning_rate": 9.404626193961427e-06,
|
| 645 |
+
"loss": 0.6186,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.24422973698336017,
|
| 650 |
+
"grad_norm": 2.719745903154331,
|
| 651 |
+
"learning_rate": 9.382255625016527e-06,
|
| 652 |
+
"loss": 0.6312,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24691358024691357,
|
| 657 |
+
"grad_norm": 2.8633960737686164,
|
| 658 |
+
"learning_rate": 9.359500122088511e-06,
|
| 659 |
+
"loss": 0.6158,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.249597423510467,
|
| 664 |
+
"grad_norm": 2.4972600338410573,
|
| 665 |
+
"learning_rate": 9.336361684003353e-06,
|
| 666 |
+
"loss": 0.629,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2522812667740204,
|
| 671 |
+
"grad_norm": 2.2735215195501697,
|
| 672 |
+
"learning_rate": 9.312842343223764e-06,
|
| 673 |
+
"loss": 0.6106,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2549651100375738,
|
| 678 |
+
"grad_norm": 2.365141838824348,
|
| 679 |
+
"learning_rate": 9.288944165670651e-06,
|
| 680 |
+
"loss": 0.6059,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2576489533011272,
|
| 685 |
+
"grad_norm": 2.5760090652843046,
|
| 686 |
+
"learning_rate": 9.264669250541658e-06,
|
| 687 |
+
"loss": 0.6246,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2603327965646806,
|
| 692 |
+
"grad_norm": 2.5329655943720226,
|
| 693 |
+
"learning_rate": 9.240019730126764e-06,
|
| 694 |
+
"loss": 0.606,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.26301663982823403,
|
| 699 |
+
"grad_norm": 2.5738935216730003,
|
| 700 |
+
"learning_rate": 9.214997769620998e-06,
|
| 701 |
+
"loss": 0.6143,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.26570048309178745,
|
| 706 |
+
"grad_norm": 2.63987142371882,
|
| 707 |
+
"learning_rate": 9.189605566934235e-06,
|
| 708 |
+
"loss": 0.602,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2683843263553409,
|
| 713 |
+
"grad_norm": 2.322948855489504,
|
| 714 |
+
"learning_rate": 9.163845352498141e-06,
|
| 715 |
+
"loss": 0.5985,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.27106816961889424,
|
| 720 |
+
"grad_norm": 2.402976520275366,
|
| 721 |
+
"learning_rate": 9.137719389070259e-06,
|
| 722 |
+
"loss": 0.608,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.27375201288244766,
|
| 727 |
+
"grad_norm": 2.456701559661602,
|
| 728 |
+
"learning_rate": 9.111229971535231e-06,
|
| 729 |
+
"loss": 0.6084,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2764358561460011,
|
| 734 |
+
"grad_norm": 2.396999722003953,
|
| 735 |
+
"learning_rate": 9.084379426703245e-06,
|
| 736 |
+
"loss": 0.6078,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.2791196994095545,
|
| 741 |
+
"grad_norm": 2.2944656877855123,
|
| 742 |
+
"learning_rate": 9.057170113105622e-06,
|
| 743 |
+
"loss": 0.5982,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.28180354267310787,
|
| 748 |
+
"grad_norm": 2.5488002762479773,
|
| 749 |
+
"learning_rate": 9.029604420787666e-06,
|
| 750 |
+
"loss": 0.595,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2844873859366613,
|
| 755 |
+
"grad_norm": 2.5621814193544656,
|
| 756 |
+
"learning_rate": 9.001684771098709e-06,
|
| 757 |
+
"loss": 0.5849,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2871712292002147,
|
| 762 |
+
"grad_norm": 2.4681234501326927,
|
| 763 |
+
"learning_rate": 8.973413616479429e-06,
|
| 764 |
+
"loss": 0.5943,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2898550724637681,
|
| 769 |
+
"grad_norm": 2.722213219660786,
|
| 770 |
+
"learning_rate": 8.944793440246435e-06,
|
| 771 |
+
"loss": 0.6,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.29253891572732155,
|
| 776 |
+
"grad_norm": 2.482994818215236,
|
| 777 |
+
"learning_rate": 8.915826756374118e-06,
|
| 778 |
+
"loss": 0.6043,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.2952227589908749,
|
| 783 |
+
"grad_norm": 2.3130329135558867,
|
| 784 |
+
"learning_rate": 8.88651610927384e-06,
|
| 785 |
+
"loss": 0.585,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.29790660225442833,
|
| 790 |
+
"grad_norm": 2.608504562134939,
|
| 791 |
+
"learning_rate": 8.856864073570429e-06,
|
| 792 |
+
"loss": 0.5769,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.30059044551798175,
|
| 797 |
+
"grad_norm": 2.518035685353769,
|
| 798 |
+
"learning_rate": 8.82687325387603e-06,
|
| 799 |
+
"loss": 0.5684,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3032742887815352,
|
| 804 |
+
"grad_norm": 2.3907513070004662,
|
| 805 |
+
"learning_rate": 8.796546284561307e-06,
|
| 806 |
+
"loss": 0.5925,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3059581320450886,
|
| 811 |
+
"grad_norm": 2.3102125436889347,
|
| 812 |
+
"learning_rate": 8.765885829524059e-06,
|
| 813 |
+
"loss": 0.5875,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.30864197530864196,
|
| 818 |
+
"grad_norm": 2.3679548171728224,
|
| 819 |
+
"learning_rate": 8.734894581955208e-06,
|
| 820 |
+
"loss": 0.5935,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3113258185721954,
|
| 825 |
+
"grad_norm": 2.384663332405018,
|
| 826 |
+
"learning_rate": 8.703575264102245e-06,
|
| 827 |
+
"loss": 0.587,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3140096618357488,
|
| 832 |
+
"grad_norm": 2.4842753819177177,
|
| 833 |
+
"learning_rate": 8.6719306270301e-06,
|
| 834 |
+
"loss": 0.5877,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3166935050993022,
|
| 839 |
+
"grad_norm": 2.4604537301515355,
|
| 840 |
+
"learning_rate": 8.639963450379494e-06,
|
| 841 |
+
"loss": 0.5937,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3193773483628556,
|
| 846 |
+
"grad_norm": 2.3929768937522464,
|
| 847 |
+
"learning_rate": 8.607676542122782e-06,
|
| 848 |
+
"loss": 0.5873,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.322061191626409,
|
| 853 |
+
"grad_norm": 2.2712444923217303,
|
| 854 |
+
"learning_rate": 8.575072738317297e-06,
|
| 855 |
+
"loss": 0.5867,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3247450348899624,
|
| 860 |
+
"grad_norm": 2.4285841734633054,
|
| 861 |
+
"learning_rate": 8.542154902856232e-06,
|
| 862 |
+
"loss": 0.5798,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.32742887815351585,
|
| 867 |
+
"grad_norm": 2.5643185358166165,
|
| 868 |
+
"learning_rate": 8.508925927217083e-06,
|
| 869 |
+
"loss": 0.5696,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.33011272141706927,
|
| 874 |
+
"grad_norm": 2.480969569279566,
|
| 875 |
+
"learning_rate": 8.475388730207662e-06,
|
| 876 |
+
"loss": 0.5531,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.33279656468062263,
|
| 881 |
+
"grad_norm": 2.4832540196963455,
|
| 882 |
+
"learning_rate": 8.441546257709708e-06,
|
| 883 |
+
"loss": 0.5653,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.33548040794417605,
|
| 888 |
+
"grad_norm": 2.3354177059858623,
|
| 889 |
+
"learning_rate": 8.407401482420129e-06,
|
| 890 |
+
"loss": 0.5788,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.33816425120772947,
|
| 895 |
+
"grad_norm": 2.542916086554623,
|
| 896 |
+
"learning_rate": 8.372957403589873e-06,
|
| 897 |
+
"loss": 0.5626,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3408480944712829,
|
| 902 |
+
"grad_norm": 2.469264768397056,
|
| 903 |
+
"learning_rate": 8.33821704676049e-06,
|
| 904 |
+
"loss": 0.5641,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3435319377348363,
|
| 909 |
+
"grad_norm": 2.335932114462721,
|
| 910 |
+
"learning_rate": 8.303183463498357e-06,
|
| 911 |
+
"loss": 0.5588,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3462157809983897,
|
| 916 |
+
"grad_norm": 2.3426121287532506,
|
| 917 |
+
"learning_rate": 8.26785973112664e-06,
|
| 918 |
+
"loss": 0.5665,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3488996242619431,
|
| 923 |
+
"grad_norm": 2.5085502164744407,
|
| 924 |
+
"learning_rate": 8.232248952454978e-06,
|
| 925 |
+
"loss": 0.5551,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3515834675254965,
|
| 930 |
+
"grad_norm": 2.367908978354936,
|
| 931 |
+
"learning_rate": 8.196354255506937e-06,
|
| 932 |
+
"loss": 0.5841,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.35426731078904994,
|
| 937 |
+
"grad_norm": 2.368147432676988,
|
| 938 |
+
"learning_rate": 8.160178793245254e-06,
|
| 939 |
+
"loss": 0.5543,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3569511540526033,
|
| 944 |
+
"grad_norm": 2.4274434068510464,
|
| 945 |
+
"learning_rate": 8.12372574329487e-06,
|
| 946 |
+
"loss": 0.553,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3596349973161567,
|
| 951 |
+
"grad_norm": 2.42778146542873,
|
| 952 |
+
"learning_rate": 8.086998307663815e-06,
|
| 953 |
+
"loss": 0.5613,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.36231884057971014,
|
| 958 |
+
"grad_norm": 2.38111212352637,
|
| 959 |
+
"learning_rate": 8.049999712461956e-06,
|
| 960 |
+
"loss": 0.5437,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.36500268384326356,
|
| 965 |
+
"grad_norm": 2.33473536282302,
|
| 966 |
+
"learning_rate": 8.012733207617602e-06,
|
| 967 |
+
"loss": 0.5519,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.367686527106817,
|
| 972 |
+
"grad_norm": 2.260812226426273,
|
| 973 |
+
"learning_rate": 7.975202066592038e-06,
|
| 974 |
+
"loss": 0.5453,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.37037037037037035,
|
| 979 |
+
"grad_norm": 2.463629677104882,
|
| 980 |
+
"learning_rate": 7.937409586091996e-06,
|
| 981 |
+
"loss": 0.552,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.37305421363392377,
|
| 986 |
+
"grad_norm": 2.507311462473454,
|
| 987 |
+
"learning_rate": 7.899359085780062e-06,
|
| 988 |
+
"loss": 0.5482,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3757380568974772,
|
| 993 |
+
"grad_norm": 2.4632736487546905,
|
| 994 |
+
"learning_rate": 7.86105390798309e-06,
|
| 995 |
+
"loss": 0.5416,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3784219001610306,
|
| 1000 |
+
"grad_norm": 2.4592304831976333,
|
| 1001 |
+
"learning_rate": 7.822497417398607e-06,
|
| 1002 |
+
"loss": 0.5461,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.38110574342458403,
|
| 1007 |
+
"grad_norm": 2.3972861209417484,
|
| 1008 |
+
"learning_rate": 7.783693000799263e-06,
|
| 1009 |
+
"loss": 0.5415,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3837895866881374,
|
| 1014 |
+
"grad_norm": 2.385175750668056,
|
| 1015 |
+
"learning_rate": 7.744644066735335e-06,
|
| 1016 |
+
"loss": 0.529,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.3864734299516908,
|
| 1021 |
+
"grad_norm": 2.2941982595733634,
|
| 1022 |
+
"learning_rate": 7.705354045235335e-06,
|
| 1023 |
+
"loss": 0.5263,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.38915727321524424,
|
| 1028 |
+
"grad_norm": 2.4098039236931217,
|
| 1029 |
+
"learning_rate": 7.665826387504708e-06,
|
| 1030 |
+
"loss": 0.5367,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.39184111647879766,
|
| 1035 |
+
"grad_norm": 2.369548787017956,
|
| 1036 |
+
"learning_rate": 7.626064565622684e-06,
|
| 1037 |
+
"loss": 0.5492,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.394524959742351,
|
| 1042 |
+
"grad_norm": 2.485485577179457,
|
| 1043 |
+
"learning_rate": 7.586072072237291e-06,
|
| 1044 |
+
"loss": 0.549,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.39720880300590444,
|
| 1049 |
+
"grad_norm": 2.3413134338829678,
|
| 1050 |
+
"learning_rate": 7.545852420258566e-06,
|
| 1051 |
+
"loss": 0.5416,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.39989264626945786,
|
| 1056 |
+
"grad_norm": 2.4643129883436163,
|
| 1057 |
+
"learning_rate": 7.505409142549987e-06,
|
| 1058 |
+
"loss": 0.5317,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4025764895330113,
|
| 1063 |
+
"grad_norm": 2.4296220280963796,
|
| 1064 |
+
"learning_rate": 7.464745791618138e-06,
|
| 1065 |
+
"loss": 0.517,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4052603327965647,
|
| 1070 |
+
"grad_norm": 2.2974621401486988,
|
| 1071 |
+
"learning_rate": 7.423865939300674e-06,
|
| 1072 |
+
"loss": 0.5291,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.40794417606011807,
|
| 1077 |
+
"grad_norm": 2.3986283475814685,
|
| 1078 |
+
"learning_rate": 7.382773176452561e-06,
|
| 1079 |
+
"loss": 0.5143,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4106280193236715,
|
| 1084 |
+
"grad_norm": 2.4550380944787893,
|
| 1085 |
+
"learning_rate": 7.341471112630667e-06,
|
| 1086 |
+
"loss": 0.5303,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4133118625872249,
|
| 1091 |
+
"grad_norm": 2.439286617193026,
|
| 1092 |
+
"learning_rate": 7.2999633757766956e-06,
|
| 1093 |
+
"loss": 0.5209,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.41599570585077833,
|
| 1098 |
+
"grad_norm": 2.579280240608442,
|
| 1099 |
+
"learning_rate": 7.258253611898509e-06,
|
| 1100 |
+
"loss": 0.5181,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.41867954911433175,
|
| 1105 |
+
"grad_norm": 2.4459646375459707,
|
| 1106 |
+
"learning_rate": 7.216345484749876e-06,
|
| 1107 |
+
"loss": 0.5285,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4213633923778851,
|
| 1112 |
+
"grad_norm": 2.31347148751011,
|
| 1113 |
+
"learning_rate": 7.174242675508636e-06,
|
| 1114 |
+
"loss": 0.5209,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.42404723564143854,
|
| 1119 |
+
"grad_norm": 2.2544720712727306,
|
| 1120 |
+
"learning_rate": 7.131948882453361e-06,
|
| 1121 |
+
"loss": 0.5189,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.42673107890499196,
|
| 1126 |
+
"grad_norm": 2.4273603418815353,
|
| 1127 |
+
"learning_rate": 7.089467820638491e-06,
|
| 1128 |
+
"loss": 0.5185,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4294149221685454,
|
| 1133 |
+
"grad_norm": 2.474461554762193,
|
| 1134 |
+
"learning_rate": 7.0468032215680116e-06,
|
| 1135 |
+
"loss": 0.5005,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.43209876543209874,
|
| 1140 |
+
"grad_norm": 2.3040840560842515,
|
| 1141 |
+
"learning_rate": 7.003958832867681e-06,
|
| 1142 |
+
"loss": 0.5089,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.43478260869565216,
|
| 1147 |
+
"grad_norm": 2.3521717336924923,
|
| 1148 |
+
"learning_rate": 6.960938417955841e-06,
|
| 1149 |
+
"loss": 0.4912,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4374664519592056,
|
| 1154 |
+
"grad_norm": 2.645750501936123,
|
| 1155 |
+
"learning_rate": 6.917745755712839e-06,
|
| 1156 |
+
"loss": 0.5103,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.440150295222759,
|
| 1161 |
+
"grad_norm": 2.3864966496000775,
|
| 1162 |
+
"learning_rate": 6.874384640149101e-06,
|
| 1163 |
+
"loss": 0.5173,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4428341384863124,
|
| 1168 |
+
"grad_norm": 2.4243382348421023,
|
| 1169 |
+
"learning_rate": 6.830858880071859e-06,
|
| 1170 |
+
"loss": 0.5008,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4455179817498658,
|
| 1175 |
+
"grad_norm": 2.3826616991498994,
|
| 1176 |
+
"learning_rate": 6.7871722987505975e-06,
|
| 1177 |
+
"loss": 0.4975,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4482018250134192,
|
| 1182 |
+
"grad_norm": 2.350730267510358,
|
| 1183 |
+
"learning_rate": 6.743328733581211e-06,
|
| 1184 |
+
"loss": 0.4974,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.45088566827697263,
|
| 1189 |
+
"grad_norm": 2.4759950761731533,
|
| 1190 |
+
"learning_rate": 6.6993320357489396e-06,
|
| 1191 |
+
"loss": 0.5109,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.45356951154052605,
|
| 1196 |
+
"grad_norm": 2.435108148176811,
|
| 1197 |
+
"learning_rate": 6.6551860698900804e-06,
|
| 1198 |
+
"loss": 0.5109,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.4562533548040794,
|
| 1203 |
+
"grad_norm": 2.328924887512505,
|
| 1204 |
+
"learning_rate": 6.610894713752513e-06,
|
| 1205 |
+
"loss": 0.511,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.45893719806763283,
|
| 1210 |
+
"grad_norm": 2.7752731374751853,
|
| 1211 |
+
"learning_rate": 6.566461857855096e-06,
|
| 1212 |
+
"loss": 0.482,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.46162104133118625,
|
| 1217 |
+
"grad_norm": 2.7298504650341933,
|
| 1218 |
+
"learning_rate": 6.52189140514591e-06,
|
| 1219 |
+
"loss": 0.4899,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4643048845947397,
|
| 1224 |
+
"grad_norm": 2.3730194932640716,
|
| 1225 |
+
"learning_rate": 6.477187270659444e-06,
|
| 1226 |
+
"loss": 0.4933,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4669887278582931,
|
| 1231 |
+
"grad_norm": 2.5257680480017943,
|
| 1232 |
+
"learning_rate": 6.432353381172686e-06,
|
| 1233 |
+
"loss": 0.5118,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.46967257112184646,
|
| 1238 |
+
"grad_norm": 2.3277533281506684,
|
| 1239 |
+
"learning_rate": 6.387393674860205e-06,
|
| 1240 |
+
"loss": 0.4726,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4723564143853999,
|
| 1245 |
+
"grad_norm": 2.360810322256424,
|
| 1246 |
+
"learning_rate": 6.3423121009482226e-06,
|
| 1247 |
+
"loss": 0.4923,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4750402576489533,
|
| 1252 |
+
"grad_norm": 2.3485189902814594,
|
| 1253 |
+
"learning_rate": 6.297112619367721e-06,
|
| 1254 |
+
"loss": 0.4812,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4777241009125067,
|
| 1259 |
+
"grad_norm": 2.4376276616820047,
|
| 1260 |
+
"learning_rate": 6.2517992004066e-06,
|
| 1261 |
+
"loss": 0.486,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.48040794417606014,
|
| 1266 |
+
"grad_norm": 2.3269791738701753,
|
| 1267 |
+
"learning_rate": 6.2063758243609275e-06,
|
| 1268 |
+
"loss": 0.5026,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.4830917874396135,
|
| 1273 |
+
"grad_norm": 2.305172802716132,
|
| 1274 |
+
"learning_rate": 6.160846481185326e-06,
|
| 1275 |
+
"loss": 0.492,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4857756307031669,
|
| 1280 |
+
"grad_norm": 2.355738561421812,
|
| 1281 |
+
"learning_rate": 6.115215170142481e-06,
|
| 1282 |
+
"loss": 0.4864,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.48845947396672035,
|
| 1287 |
+
"grad_norm": 2.2772567999964894,
|
| 1288 |
+
"learning_rate": 6.069485899451863e-06,
|
| 1289 |
+
"loss": 0.4957,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.49114331723027377,
|
| 1294 |
+
"grad_norm": 2.293960156110535,
|
| 1295 |
+
"learning_rate": 6.023662685937643e-06,
|
| 1296 |
+
"loss": 0.4785,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.49382716049382713,
|
| 1301 |
+
"grad_norm": 2.452682027740044,
|
| 1302 |
+
"learning_rate": 5.9777495546758556e-06,
|
| 1303 |
+
"loss": 0.4706,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.49651100375738055,
|
| 1308 |
+
"grad_norm": 2.3356554581318307,
|
| 1309 |
+
"learning_rate": 5.9317505386408436e-06,
|
| 1310 |
+
"loss": 0.4879,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.499194847020934,
|
| 1315 |
+
"grad_norm": 2.413325314490582,
|
| 1316 |
+
"learning_rate": 5.885669678350997e-06,
|
| 1317 |
+
"loss": 0.4836,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5018786902844874,
|
| 1322 |
+
"grad_norm": 2.34645866757674,
|
| 1323 |
+
"learning_rate": 5.839511021513853e-06,
|
| 1324 |
+
"loss": 0.4732,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5045625335480408,
|
| 1329 |
+
"grad_norm": 2.2963428361302904,
|
| 1330 |
+
"learning_rate": 5.793278622670527e-06,
|
| 1331 |
+
"loss": 0.4768,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5072463768115942,
|
| 1336 |
+
"grad_norm": 2.365796624292683,
|
| 1337 |
+
"learning_rate": 5.746976542839583e-06,
|
| 1338 |
+
"loss": 0.4723,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5099302200751477,
|
| 1343 |
+
"grad_norm": 2.2909468487010347,
|
| 1344 |
+
"learning_rate": 5.700608849160307e-06,
|
| 1345 |
+
"loss": 0.4632,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.512614063338701,
|
| 1350 |
+
"grad_norm": 2.400437273534213,
|
| 1351 |
+
"learning_rate": 5.654179614535457e-06,
|
| 1352 |
+
"loss": 0.4558,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5152979066022544,
|
| 1357 |
+
"grad_norm": 2.366575335205731,
|
| 1358 |
+
"learning_rate": 5.607692917273499e-06,
|
| 1359 |
+
"loss": 0.4651,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5179817498658078,
|
| 1364 |
+
"grad_norm": 2.2766517969507554,
|
| 1365 |
+
"learning_rate": 5.561152840730371e-06,
|
| 1366 |
+
"loss": 0.4559,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5206655931293612,
|
| 1371 |
+
"grad_norm": 2.2683862531318733,
|
| 1372 |
+
"learning_rate": 5.514563472950812e-06,
|
| 1373 |
+
"loss": 0.4501,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5233494363929146,
|
| 1378 |
+
"grad_norm": 2.307175024358814,
|
| 1379 |
+
"learning_rate": 5.46792890630926e-06,
|
| 1380 |
+
"loss": 0.451,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5260332796564681,
|
| 1385 |
+
"grad_norm": 2.3931220819772174,
|
| 1386 |
+
"learning_rate": 5.421253237150389e-06,
|
| 1387 |
+
"loss": 0.4562,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5287171229200215,
|
| 1392 |
+
"grad_norm": 2.230895258923371,
|
| 1393 |
+
"learning_rate": 5.374540565429288e-06,
|
| 1394 |
+
"loss": 0.4703,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5314009661835749,
|
| 1399 |
+
"grad_norm": 2.3419568661003147,
|
| 1400 |
+
"learning_rate": 5.327794994351321e-06,
|
| 1401 |
+
"loss": 0.4641,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5340848094471283,
|
| 1406 |
+
"grad_norm": 2.304734852864055,
|
| 1407 |
+
"learning_rate": 5.281020630011703e-06,
|
| 1408 |
+
"loss": 0.463,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5367686527106817,
|
| 1413 |
+
"grad_norm": 2.2318721628819413,
|
| 1414 |
+
"learning_rate": 5.234221581034834e-06,
|
| 1415 |
+
"loss": 0.4541,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5394524959742351,
|
| 1420 |
+
"grad_norm": 2.3906459933439077,
|
| 1421 |
+
"learning_rate": 5.187401958213391e-06,
|
| 1422 |
+
"loss": 0.4525,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5421363392377885,
|
| 1427 |
+
"grad_norm": 2.467365672509074,
|
| 1428 |
+
"learning_rate": 5.140565874147236e-06,
|
| 1429 |
+
"loss": 0.4501,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5448201825013419,
|
| 1434 |
+
"grad_norm": 2.277483626055024,
|
| 1435 |
+
"learning_rate": 5.093717442882185e-06,
|
| 1436 |
+
"loss": 0.4545,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5475040257648953,
|
| 1441 |
+
"grad_norm": 2.3593150665128393,
|
| 1442 |
+
"learning_rate": 5.046860779548613e-06,
|
| 1443 |
+
"loss": 0.4483,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5501878690284487,
|
| 1448 |
+
"grad_norm": 2.258246761530304,
|
| 1449 |
+
"learning_rate": 5e-06,
|
| 1450 |
+
"loss": 0.4546,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5528717122920022,
|
| 1455 |
+
"grad_norm": 2.370134681916226,
|
| 1456 |
+
"learning_rate": 4.953139220451386e-06,
|
| 1457 |
+
"loss": 0.4515,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5555555555555556,
|
| 1462 |
+
"grad_norm": 2.2408497142914263,
|
| 1463 |
+
"learning_rate": 4.906282557117817e-06,
|
| 1464 |
+
"loss": 0.4341,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.558239398819109,
|
| 1469 |
+
"grad_norm": 2.3069571005115286,
|
| 1470 |
+
"learning_rate": 4.859434125852764e-06,
|
| 1471 |
+
"loss": 0.443,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5609232420826624,
|
| 1476 |
+
"grad_norm": 2.3320315085107963,
|
| 1477 |
+
"learning_rate": 4.812598041786611e-06,
|
| 1478 |
+
"loss": 0.4394,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5636070853462157,
|
| 1483 |
+
"grad_norm": 2.3442024150109355,
|
| 1484 |
+
"learning_rate": 4.765778418965168e-06,
|
| 1485 |
+
"loss": 0.4457,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5662909286097692,
|
| 1490 |
+
"grad_norm": 2.3358028069710355,
|
| 1491 |
+
"learning_rate": 4.718979369988299e-06,
|
| 1492 |
+
"loss": 0.4391,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5689747718733226,
|
| 1497 |
+
"grad_norm": 2.5507352210470993,
|
| 1498 |
+
"learning_rate": 4.672205005648681e-06,
|
| 1499 |
+
"loss": 0.437,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.571658615136876,
|
| 1504 |
+
"grad_norm": 2.309839111983691,
|
| 1505 |
+
"learning_rate": 4.625459434570712e-06,
|
| 1506 |
+
"loss": 0.4412,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5743424584004294,
|
| 1511 |
+
"grad_norm": 2.3271635616011768,
|
| 1512 |
+
"learning_rate": 4.578746762849612e-06,
|
| 1513 |
+
"loss": 0.4494,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5770263016639828,
|
| 1518 |
+
"grad_norm": 2.3546579002361514,
|
| 1519 |
+
"learning_rate": 4.532071093690741e-06,
|
| 1520 |
+
"loss": 0.4379,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5797101449275363,
|
| 1525 |
+
"grad_norm": 2.2413776268605665,
|
| 1526 |
+
"learning_rate": 4.485436527049189e-06,
|
| 1527 |
+
"loss": 0.432,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5823939881910897,
|
| 1532 |
+
"grad_norm": 2.1798039451185613,
|
| 1533 |
+
"learning_rate": 4.438847159269631e-06,
|
| 1534 |
+
"loss": 0.4258,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5850778314546431,
|
| 1539 |
+
"grad_norm": 2.3359688865498662,
|
| 1540 |
+
"learning_rate": 4.392307082726503e-06,
|
| 1541 |
+
"loss": 0.4359,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5877616747181964,
|
| 1546 |
+
"grad_norm": 2.3364533232377998,
|
| 1547 |
+
"learning_rate": 4.345820385464543e-06,
|
| 1548 |
+
"loss": 0.4281,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5904455179817498,
|
| 1553 |
+
"grad_norm": 2.3860711291519356,
|
| 1554 |
+
"learning_rate": 4.299391150839694e-06,
|
| 1555 |
+
"loss": 0.4364,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5931293612453032,
|
| 1560 |
+
"grad_norm": 2.276752067399093,
|
| 1561 |
+
"learning_rate": 4.253023457160418e-06,
|
| 1562 |
+
"loss": 0.4186,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5958132045088567,
|
| 1567 |
+
"grad_norm": 2.467553896688473,
|
| 1568 |
+
"learning_rate": 4.2067213773294745e-06,
|
| 1569 |
+
"loss": 0.4387,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5984970477724101,
|
| 1574 |
+
"grad_norm": 2.3540003775384046,
|
| 1575 |
+
"learning_rate": 4.16048897848615e-06,
|
| 1576 |
+
"loss": 0.43,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6011808910359635,
|
| 1581 |
+
"grad_norm": 2.505862126720908,
|
| 1582 |
+
"learning_rate": 4.114330321649004e-06,
|
| 1583 |
+
"loss": 0.4321,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6038647342995169,
|
| 1588 |
+
"grad_norm": 2.4630371968354656,
|
| 1589 |
+
"learning_rate": 4.068249461359158e-06,
|
| 1590 |
+
"loss": 0.4297,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6065485775630703,
|
| 1595 |
+
"grad_norm": 2.272836354134841,
|
| 1596 |
+
"learning_rate": 4.022250445324145e-06,
|
| 1597 |
+
"loss": 0.4268,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6092324208266238,
|
| 1602 |
+
"grad_norm": 2.2886295946366975,
|
| 1603 |
+
"learning_rate": 3.976337314062358e-06,
|
| 1604 |
+
"loss": 0.4167,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6119162640901772,
|
| 1609 |
+
"grad_norm": 2.2103585092808617,
|
| 1610 |
+
"learning_rate": 3.9305141005481385e-06,
|
| 1611 |
+
"loss": 0.4199,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6146001073537305,
|
| 1616 |
+
"grad_norm": 2.2740765745796123,
|
| 1617 |
+
"learning_rate": 3.8847848298575205e-06,
|
| 1618 |
+
"loss": 0.4189,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6172839506172839,
|
| 1623 |
+
"grad_norm": 2.258557673733666,
|
| 1624 |
+
"learning_rate": 3.839153518814677e-06,
|
| 1625 |
+
"loss": 0.4306,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6199677938808373,
|
| 1630 |
+
"grad_norm": 2.2068268474418358,
|
| 1631 |
+
"learning_rate": 3.7936241756390746e-06,
|
| 1632 |
+
"loss": 0.4247,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6226516371443908,
|
| 1637 |
+
"grad_norm": 2.194022472389834,
|
| 1638 |
+
"learning_rate": 3.748200799593402e-06,
|
| 1639 |
+
"loss": 0.4231,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6253354804079442,
|
| 1644 |
+
"grad_norm": 2.326988084434494,
|
| 1645 |
+
"learning_rate": 3.7028873806322797e-06,
|
| 1646 |
+
"loss": 0.4238,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6280193236714976,
|
| 1651 |
+
"grad_norm": 2.2517958835294296,
|
| 1652 |
+
"learning_rate": 3.6576878990517783e-06,
|
| 1653 |
+
"loss": 0.416,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.630703166935051,
|
| 1658 |
+
"grad_norm": 2.1808207196710883,
|
| 1659 |
+
"learning_rate": 3.6126063251397968e-06,
|
| 1660 |
+
"loss": 0.4287,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6333870101986044,
|
| 1665 |
+
"grad_norm": 2.3042249504814047,
|
| 1666 |
+
"learning_rate": 3.567646618827315e-06,
|
| 1667 |
+
"loss": 0.4125,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6360708534621579,
|
| 1672 |
+
"grad_norm": 2.3472786605740343,
|
| 1673 |
+
"learning_rate": 3.522812729340557e-06,
|
| 1674 |
+
"loss": 0.4206,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6387546967257112,
|
| 1679 |
+
"grad_norm": 2.2475547544387866,
|
| 1680 |
+
"learning_rate": 3.4781085948540907e-06,
|
| 1681 |
+
"loss": 0.398,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6414385399892646,
|
| 1686 |
+
"grad_norm": 2.287893454196299,
|
| 1687 |
+
"learning_rate": 3.4335381421449056e-06,
|
| 1688 |
+
"loss": 0.4078,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.644122383252818,
|
| 1693 |
+
"grad_norm": 2.3938598824979236,
|
| 1694 |
+
"learning_rate": 3.3891052862474884e-06,
|
| 1695 |
+
"loss": 0.4184,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6468062265163714,
|
| 1700 |
+
"grad_norm": 2.284784413306213,
|
| 1701 |
+
"learning_rate": 3.3448139301099212e-06,
|
| 1702 |
+
"loss": 0.413,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6494900697799249,
|
| 1707 |
+
"grad_norm": 2.313513606221783,
|
| 1708 |
+
"learning_rate": 3.3006679642510596e-06,
|
| 1709 |
+
"loss": 0.4222,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6521739130434783,
|
| 1714 |
+
"grad_norm": 2.2811158566992407,
|
| 1715 |
+
"learning_rate": 3.2566712664187907e-06,
|
| 1716 |
+
"loss": 0.4103,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6548577563070317,
|
| 1721 |
+
"grad_norm": 2.231261781303834,
|
| 1722 |
+
"learning_rate": 3.2128277012494046e-06,
|
| 1723 |
+
"loss": 0.4073,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6575415995705851,
|
| 1728 |
+
"grad_norm": 2.377033426744753,
|
| 1729 |
+
"learning_rate": 3.169141119928142e-06,
|
| 1730 |
+
"loss": 0.4083,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6602254428341385,
|
| 1735 |
+
"grad_norm": 2.2511272231711787,
|
| 1736 |
+
"learning_rate": 3.125615359850902e-06,
|
| 1737 |
+
"loss": 0.3908,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6629092860976918,
|
| 1742 |
+
"grad_norm": 2.1422423919919233,
|
| 1743 |
+
"learning_rate": 3.082254244287163e-06,
|
| 1744 |
+
"loss": 0.3956,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6655931293612453,
|
| 1749 |
+
"grad_norm": 2.1833809439258007,
|
| 1750 |
+
"learning_rate": 3.0390615820441605e-06,
|
| 1751 |
+
"loss": 0.4078,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6682769726247987,
|
| 1756 |
+
"grad_norm": 2.0915894234273305,
|
| 1757 |
+
"learning_rate": 2.9960411671323215e-06,
|
| 1758 |
+
"loss": 0.415,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6709608158883521,
|
| 1763 |
+
"grad_norm": 2.3267272756698767,
|
| 1764 |
+
"learning_rate": 2.9531967784319897e-06,
|
| 1765 |
+
"loss": 0.3998,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6736446591519055,
|
| 1770 |
+
"grad_norm": 2.2454024145911147,
|
| 1771 |
+
"learning_rate": 2.9105321793615106e-06,
|
| 1772 |
+
"loss": 0.4139,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6763285024154589,
|
| 1777 |
+
"grad_norm": 2.1672873136892337,
|
| 1778 |
+
"learning_rate": 2.8680511175466384e-06,
|
| 1779 |
+
"loss": 0.4012,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6790123456790124,
|
| 1784 |
+
"grad_norm": 2.36164728633219,
|
| 1785 |
+
"learning_rate": 2.8257573244913637e-06,
|
| 1786 |
+
"loss": 0.4053,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6816961889425658,
|
| 1791 |
+
"grad_norm": 2.198633451961816,
|
| 1792 |
+
"learning_rate": 2.783654515250126e-06,
|
| 1793 |
+
"loss": 0.3895,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6843800322061192,
|
| 1798 |
+
"grad_norm": 2.1741967245818454,
|
| 1799 |
+
"learning_rate": 2.741746388101493e-06,
|
| 1800 |
+
"loss": 0.4037,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6870638754696726,
|
| 1805 |
+
"grad_norm": 2.4661760766296754,
|
| 1806 |
+
"learning_rate": 2.700036624223308e-06,
|
| 1807 |
+
"loss": 0.4011,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6897477187332259,
|
| 1812 |
+
"grad_norm": 2.283551766561198,
|
| 1813 |
+
"learning_rate": 2.6585288873693355e-06,
|
| 1814 |
+
"loss": 0.398,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6924315619967794,
|
| 1819 |
+
"grad_norm": 2.2282092986087796,
|
| 1820 |
+
"learning_rate": 2.6172268235474396e-06,
|
| 1821 |
+
"loss": 0.3896,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6951154052603328,
|
| 1826 |
+
"grad_norm": 2.190285384279946,
|
| 1827 |
+
"learning_rate": 2.576134060699328e-06,
|
| 1828 |
+
"loss": 0.4081,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6977992485238862,
|
| 1833 |
+
"grad_norm": 2.1692244547778405,
|
| 1834 |
+
"learning_rate": 2.5352542083818636e-06,
|
| 1835 |
+
"loss": 0.3887,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7004830917874396,
|
| 1840 |
+
"grad_norm": 2.2586754461746126,
|
| 1841 |
+
"learning_rate": 2.494590857450016e-06,
|
| 1842 |
+
"loss": 0.3957,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.703166935050993,
|
| 1847 |
+
"grad_norm": 2.374656982408945,
|
| 1848 |
+
"learning_rate": 2.454147579741436e-06,
|
| 1849 |
+
"loss": 0.3874,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7058507783145465,
|
| 1854 |
+
"grad_norm": 2.2261704008153536,
|
| 1855 |
+
"learning_rate": 2.4139279277627113e-06,
|
| 1856 |
+
"loss": 0.388,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7085346215780999,
|
| 1861 |
+
"grad_norm": 2.2437976230745384,
|
| 1862 |
+
"learning_rate": 2.373935434377316e-06,
|
| 1863 |
+
"loss": 0.3905,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7112184648416533,
|
| 1868 |
+
"grad_norm": 2.1589184736112075,
|
| 1869 |
+
"learning_rate": 2.3341736124952924e-06,
|
| 1870 |
+
"loss": 0.3928,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7139023081052066,
|
| 1875 |
+
"grad_norm": 2.1883376186173638,
|
| 1876 |
+
"learning_rate": 2.294645954764666e-06,
|
| 1877 |
+
"loss": 0.3937,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.71658615136876,
|
| 1882 |
+
"grad_norm": 2.115522768168307,
|
| 1883 |
+
"learning_rate": 2.2553559332646675e-06,
|
| 1884 |
+
"loss": 0.3738,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7192699946323134,
|
| 1889 |
+
"grad_norm": 2.2554699265788014,
|
| 1890 |
+
"learning_rate": 2.216306999200739e-06,
|
| 1891 |
+
"loss": 0.3756,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7219538378958669,
|
| 1896 |
+
"grad_norm": 2.1331294949741513,
|
| 1897 |
+
"learning_rate": 2.1775025826013953e-06,
|
| 1898 |
+
"loss": 0.3794,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7246376811594203,
|
| 1903 |
+
"grad_norm": 2.2181208326555892,
|
| 1904 |
+
"learning_rate": 2.1389460920169105e-06,
|
| 1905 |
+
"loss": 0.3739,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7273215244229737,
|
| 1910 |
+
"grad_norm": 2.3464498796911317,
|
| 1911 |
+
"learning_rate": 2.100640914219939e-06,
|
| 1912 |
+
"loss": 0.4009,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7300053676865271,
|
| 1917 |
+
"grad_norm": 2.205656545431609,
|
| 1918 |
+
"learning_rate": 2.062590413908007e-06,
|
| 1919 |
+
"loss": 0.3926,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7326892109500805,
|
| 1924 |
+
"grad_norm": 2.1357870746338414,
|
| 1925 |
+
"learning_rate": 2.0247979334079646e-06,
|
| 1926 |
+
"loss": 0.3877,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.735373054213634,
|
| 1931 |
+
"grad_norm": 2.1907334794002744,
|
| 1932 |
+
"learning_rate": 1.9872667923823996e-06,
|
| 1933 |
+
"loss": 0.38,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7380568974771873,
|
| 1938 |
+
"grad_norm": 2.1251695806868596,
|
| 1939 |
+
"learning_rate": 1.9500002875380458e-06,
|
| 1940 |
+
"loss": 0.3933,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7407407407407407,
|
| 1945 |
+
"grad_norm": 2.1564914485662734,
|
| 1946 |
+
"learning_rate": 1.9130016923361843e-06,
|
| 1947 |
+
"loss": 0.3933,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7434245840042941,
|
| 1952 |
+
"grad_norm": 2.036260794536821,
|
| 1953 |
+
"learning_rate": 1.8762742567051318e-06,
|
| 1954 |
+
"loss": 0.3765,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7461084272678475,
|
| 1959 |
+
"grad_norm": 2.116566265103448,
|
| 1960 |
+
"learning_rate": 1.8398212067547455e-06,
|
| 1961 |
+
"loss": 0.3709,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.748792270531401,
|
| 1966 |
+
"grad_norm": 2.22514017476847,
|
| 1967 |
+
"learning_rate": 1.8036457444930643e-06,
|
| 1968 |
+
"loss": 0.3789,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7514761137949544,
|
| 1973 |
+
"grad_norm": 2.213733855156735,
|
| 1974 |
+
"learning_rate": 1.7677510475450244e-06,
|
| 1975 |
+
"loss": 0.3842,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7541599570585078,
|
| 1980 |
+
"grad_norm": 2.181820066227329,
|
| 1981 |
+
"learning_rate": 1.7321402688733618e-06,
|
| 1982 |
+
"loss": 0.3806,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7568438003220612,
|
| 1987 |
+
"grad_norm": 2.2050398353530203,
|
| 1988 |
+
"learning_rate": 1.6968165365016443e-06,
|
| 1989 |
+
"loss": 0.3823,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7595276435856146,
|
| 1994 |
+
"grad_norm": 2.19530752654402,
|
| 1995 |
+
"learning_rate": 1.661782953239512e-06,
|
| 1996 |
+
"loss": 0.3793,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7622114868491681,
|
| 2001 |
+
"grad_norm": 2.1495195721953024,
|
| 2002 |
+
"learning_rate": 1.6270425964101272e-06,
|
| 2003 |
+
"loss": 0.3741,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7648953301127214,
|
| 2008 |
+
"grad_norm": 2.103554729221664,
|
| 2009 |
+
"learning_rate": 1.5925985175798741e-06,
|
| 2010 |
+
"loss": 0.3729,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7675791733762748,
|
| 2015 |
+
"grad_norm": 2.076683083573831,
|
| 2016 |
+
"learning_rate": 1.5584537422902923e-06,
|
| 2017 |
+
"loss": 0.3801,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7702630166398282,
|
| 2022 |
+
"grad_norm": 2.058902410005422,
|
| 2023 |
+
"learning_rate": 1.5246112697923389e-06,
|
| 2024 |
+
"loss": 0.3716,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7729468599033816,
|
| 2029 |
+
"grad_norm": 2.12166477747948,
|
| 2030 |
+
"learning_rate": 1.4910740727829176e-06,
|
| 2031 |
+
"loss": 0.3736,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.775630703166935,
|
| 2036 |
+
"grad_norm": 2.0736640828083344,
|
| 2037 |
+
"learning_rate": 1.457845097143769e-06,
|
| 2038 |
+
"loss": 0.3656,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7783145464304885,
|
| 2043 |
+
"grad_norm": 2.162464954805237,
|
| 2044 |
+
"learning_rate": 1.424927261682703e-06,
|
| 2045 |
+
"loss": 0.3645,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7809983896940419,
|
| 2050 |
+
"grad_norm": 2.322539410509905,
|
| 2051 |
+
"learning_rate": 1.3923234578772177e-06,
|
| 2052 |
+
"loss": 0.3737,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7836822329575953,
|
| 2057 |
+
"grad_norm": 2.0847406596927334,
|
| 2058 |
+
"learning_rate": 1.3600365496205059e-06,
|
| 2059 |
+
"loss": 0.367,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7863660762211487,
|
| 2064 |
+
"grad_norm": 2.175126277242884,
|
| 2065 |
+
"learning_rate": 1.3280693729699018e-06,
|
| 2066 |
+
"loss": 0.3719,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.789049919484702,
|
| 2071 |
+
"grad_norm": 2.0902393466668983,
|
| 2072 |
+
"learning_rate": 1.2964247358977567e-06,
|
| 2073 |
+
"loss": 0.3657,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7917337627482555,
|
| 2078 |
+
"grad_norm": 1.9974175235545193,
|
| 2079 |
+
"learning_rate": 1.265105418044793e-06,
|
| 2080 |
+
"loss": 0.3567,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7944176060118089,
|
| 2085 |
+
"grad_norm": 2.3125749433062746,
|
| 2086 |
+
"learning_rate": 1.2341141704759418e-06,
|
| 2087 |
+
"loss": 0.3638,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7971014492753623,
|
| 2092 |
+
"grad_norm": 2.08290293868451,
|
| 2093 |
+
"learning_rate": 1.2034537154386933e-06,
|
| 2094 |
+
"loss": 0.3785,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7997852925389157,
|
| 2099 |
+
"grad_norm": 2.0481414392713626,
|
| 2100 |
+
"learning_rate": 1.173126746123972e-06,
|
| 2101 |
+
"loss": 0.3665,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8024691358024691,
|
| 2106 |
+
"grad_norm": 1.9820157935875906,
|
| 2107 |
+
"learning_rate": 1.1431359264295717e-06,
|
| 2108 |
+
"loss": 0.3613,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8051529790660226,
|
| 2113 |
+
"grad_norm": 2.2394169396929793,
|
| 2114 |
+
"learning_rate": 1.113483890726162e-06,
|
| 2115 |
+
"loss": 0.3679,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.807836822329576,
|
| 2120 |
+
"grad_norm": 2.2229175878011516,
|
| 2121 |
+
"learning_rate": 1.084173243625884e-06,
|
| 2122 |
+
"loss": 0.3515,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8105206655931294,
|
| 2127 |
+
"grad_norm": 2.1249532345306976,
|
| 2128 |
+
"learning_rate": 1.0552065597535671e-06,
|
| 2129 |
+
"loss": 0.3541,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8132045088566827,
|
| 2134 |
+
"grad_norm": 2.151379889387235,
|
| 2135 |
+
"learning_rate": 1.0265863835205709e-06,
|
| 2136 |
+
"loss": 0.3565,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8158883521202361,
|
| 2141 |
+
"grad_norm": 2.247568453760684,
|
| 2142 |
+
"learning_rate": 9.983152289012926e-07,
|
| 2143 |
+
"loss": 0.3767,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8185721953837896,
|
| 2148 |
+
"grad_norm": 2.030213334768943,
|
| 2149 |
+
"learning_rate": 9.70395579212336e-07,
|
| 2150 |
+
"loss": 0.3626,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.821256038647343,
|
| 2155 |
+
"grad_norm": 2.16307272105344,
|
| 2156 |
+
"learning_rate": 9.428298868943791e-07,
|
| 2157 |
+
"loss": 0.3661,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8239398819108964,
|
| 2162 |
+
"grad_norm": 1.9353846707087516,
|
| 2163 |
+
"learning_rate": 9.15620573296756e-07,
|
| 2164 |
+
"loss": 0.3559,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8266237251744498,
|
| 2169 |
+
"grad_norm": 2.4706388638021948,
|
| 2170 |
+
"learning_rate": 8.887700284647699e-07,
|
| 2171 |
+
"loss": 0.3502,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8293075684380032,
|
| 2176 |
+
"grad_norm": 2.0964124856887505,
|
| 2177 |
+
"learning_rate": 8.622806109297432e-07,
|
| 2178 |
+
"loss": 0.3635,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8319914117015567,
|
| 2183 |
+
"grad_norm": 2.1655642607273213,
|
| 2184 |
+
"learning_rate": 8.361546475018589e-07,
|
| 2185 |
+
"loss": 0.3524,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8346752549651101,
|
| 2190 |
+
"grad_norm": 2.151020178624645,
|
| 2191 |
+
"learning_rate": 8.103944330657665e-07,
|
| 2192 |
+
"loss": 0.3631,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8373590982286635,
|
| 2197 |
+
"grad_norm": 2.2274203854683643,
|
| 2198 |
+
"learning_rate": 7.850022303790033e-07,
|
| 2199 |
+
"loss": 0.3536,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8400429414922168,
|
| 2204 |
+
"grad_norm": 2.0005675417169084,
|
| 2205 |
+
"learning_rate": 7.599802698732356e-07,
|
| 2206 |
+
"loss": 0.3504,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8427267847557702,
|
| 2211 |
+
"grad_norm": 2.048544707751455,
|
| 2212 |
+
"learning_rate": 7.353307494583445e-07,
|
| 2213 |
+
"loss": 0.3688,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8454106280193237,
|
| 2218 |
+
"grad_norm": 2.286454488879537,
|
| 2219 |
+
"learning_rate": 7.1105583432935e-07,
|
| 2220 |
+
"loss": 0.3563,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8480944712828771,
|
| 2225 |
+
"grad_norm": 2.1209732067930878,
|
| 2226 |
+
"learning_rate": 6.871576567762373e-07,
|
| 2227 |
+
"loss": 0.3481,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8507783145464305,
|
| 2232 |
+
"grad_norm": 2.0479042834863113,
|
| 2233 |
+
"learning_rate": 6.63638315996647e-07,
|
| 2234 |
+
"loss": 0.3746,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8534621578099839,
|
| 2239 |
+
"grad_norm": 2.0564571087729733,
|
| 2240 |
+
"learning_rate": 6.404998779114912e-07,
|
| 2241 |
+
"loss": 0.3599,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8561460010735373,
|
| 2246 |
+
"grad_norm": 2.110264353818972,
|
| 2247 |
+
"learning_rate": 6.177443749834743e-07,
|
| 2248 |
+
"loss": 0.3498,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8588298443370908,
|
| 2253 |
+
"grad_norm": 1.958321769357571,
|
| 2254 |
+
"learning_rate": 5.953738060385761e-07,
|
| 2255 |
+
"loss": 0.3497,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8615136876006442,
|
| 2260 |
+
"grad_norm": 2.1502748066197843,
|
| 2261 |
+
"learning_rate": 5.733901360904648e-07,
|
| 2262 |
+
"loss": 0.3571,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8641975308641975,
|
| 2267 |
+
"grad_norm": 2.3497614671533262,
|
| 2268 |
+
"learning_rate": 5.517952961678997e-07,
|
| 2269 |
+
"loss": 0.348,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8668813741277509,
|
| 2274 |
+
"grad_norm": 2.035748778996207,
|
| 2275 |
+
"learning_rate": 5.305911831451044e-07,
|
| 2276 |
+
"loss": 0.3471,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8695652173913043,
|
| 2281 |
+
"grad_norm": 2.056520258073967,
|
| 2282 |
+
"learning_rate": 5.097796595751564e-07,
|
| 2283 |
+
"loss": 0.3645,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8722490606548577,
|
| 2288 |
+
"grad_norm": 1.9264861733728513,
|
| 2289 |
+
"learning_rate": 4.89362553526368e-07,
|
| 2290 |
+
"loss": 0.3416,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8749329039184112,
|
| 2295 |
+
"grad_norm": 1.983591179837642,
|
| 2296 |
+
"learning_rate": 4.6934165842172476e-07,
|
| 2297 |
+
"loss": 0.3526,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8776167471819646,
|
| 2302 |
+
"grad_norm": 2.3439353018231053,
|
| 2303 |
+
"learning_rate": 4.4971873288134237e-07,
|
| 2304 |
+
"loss": 0.3545,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.880300590445518,
|
| 2309 |
+
"grad_norm": 2.0168435139254774,
|
| 2310 |
+
"learning_rate": 4.304955005679962e-07,
|
| 2311 |
+
"loss": 0.3585,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8829844337090714,
|
| 2316 |
+
"grad_norm": 2.0767665979805945,
|
| 2317 |
+
"learning_rate": 4.1167365003571047e-07,
|
| 2318 |
+
"loss": 0.362,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8856682769726248,
|
| 2323 |
+
"grad_norm": 2.2587982125889914,
|
| 2324 |
+
"learning_rate": 3.9325483458144684e-07,
|
| 2325 |
+
"loss": 0.3588,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8883521202361782,
|
| 2330 |
+
"grad_norm": 2.0212128576246022,
|
| 2331 |
+
"learning_rate": 3.752406720998691e-07,
|
| 2332 |
+
"loss": 0.3501,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8910359634997316,
|
| 2337 |
+
"grad_norm": 2.174607134671128,
|
| 2338 |
+
"learning_rate": 3.5763274494123424e-07,
|
| 2339 |
+
"loss": 0.3504,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.893719806763285,
|
| 2344 |
+
"grad_norm": 1.8892853467813568,
|
| 2345 |
+
"learning_rate": 3.404325997723995e-07,
|
| 2346 |
+
"loss": 0.348,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8964036500268384,
|
| 2351 |
+
"grad_norm": 2.202688784379415,
|
| 2352 |
+
"learning_rate": 3.2364174744096587e-07,
|
| 2353 |
+
"loss": 0.3474,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8990874932903918,
|
| 2358 |
+
"grad_norm": 1.9882966199629937,
|
| 2359 |
+
"learning_rate": 3.072616628425601e-07,
|
| 2360 |
+
"loss": 0.3488,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9017713365539453,
|
| 2365 |
+
"grad_norm": 2.042351450731294,
|
| 2366 |
+
"learning_rate": 2.9129378479129e-07,
|
| 2367 |
+
"loss": 0.3521,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9044551798174987,
|
| 2372 |
+
"grad_norm": 1.9810136810117025,
|
| 2373 |
+
"learning_rate": 2.7573951589335177e-07,
|
| 2374 |
+
"loss": 0.3576,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9071390230810521,
|
| 2379 |
+
"grad_norm": 2.039125294256416,
|
| 2380 |
+
"learning_rate": 2.60600222423833e-07,
|
| 2381 |
+
"loss": 0.3595,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9098228663446055,
|
| 2386 |
+
"grad_norm": 2.1305767673918568,
|
| 2387 |
+
"learning_rate": 2.45877234206694e-07,
|
| 2388 |
+
"loss": 0.3506,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9125067096081588,
|
| 2393 |
+
"grad_norm": 1.999762682379838,
|
| 2394 |
+
"learning_rate": 2.3157184449796365e-07,
|
| 2395 |
+
"loss": 0.3451,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9151905528717122,
|
| 2400 |
+
"grad_norm": 2.4121285473633343,
|
| 2401 |
+
"learning_rate": 2.1768530987213267e-07,
|
| 2402 |
+
"loss": 0.3561,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9178743961352657,
|
| 2407 |
+
"grad_norm": 2.1045057194711383,
|
| 2408 |
+
"learning_rate": 2.0421885011178532e-07,
|
| 2409 |
+
"loss": 0.3633,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9205582393988191,
|
| 2414 |
+
"grad_norm": 2.036353438280661,
|
| 2415 |
+
"learning_rate": 1.911736481004489e-07,
|
| 2416 |
+
"loss": 0.3439,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9232420826623725,
|
| 2421 |
+
"grad_norm": 2.0085164573021834,
|
| 2422 |
+
"learning_rate": 1.7855084971869385e-07,
|
| 2423 |
+
"loss": 0.3522,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9259259259259259,
|
| 2428 |
+
"grad_norm": 1.9703347585874995,
|
| 2429 |
+
"learning_rate": 1.663515637434765e-07,
|
| 2430 |
+
"loss": 0.3491,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9286097691894794,
|
| 2435 |
+
"grad_norm": 1.9548157965432644,
|
| 2436 |
+
"learning_rate": 1.5457686175074816e-07,
|
| 2437 |
+
"loss": 0.3485,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9312936124530328,
|
| 2442 |
+
"grad_norm": 2.0143091147389574,
|
| 2443 |
+
"learning_rate": 1.432277780213298e-07,
|
| 2444 |
+
"loss": 0.3438,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9339774557165862,
|
| 2449 |
+
"grad_norm": 1.9390821452634188,
|
| 2450 |
+
"learning_rate": 1.3230530945005704e-07,
|
| 2451 |
+
"loss": 0.3482,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9366612989801396,
|
| 2456 |
+
"grad_norm": 1.9503085883819737,
|
| 2457 |
+
"learning_rate": 1.2181041545821727e-07,
|
| 2458 |
+
"loss": 0.3584,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9393451422436929,
|
| 2463 |
+
"grad_norm": 2.1428772506615372,
|
| 2464 |
+
"learning_rate": 1.1174401790927325e-07,
|
| 2465 |
+
"loss": 0.3509,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9420289855072463,
|
| 2470 |
+
"grad_norm": 2.058814075788709,
|
| 2471 |
+
"learning_rate": 1.0210700102788796e-07,
|
| 2472 |
+
"loss": 0.3483,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9447128287707998,
|
| 2477 |
+
"grad_norm": 2.0643232309660924,
|
| 2478 |
+
"learning_rate": 9.290021132225546e-08,
|
| 2479 |
+
"loss": 0.3514,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9473966720343532,
|
| 2484 |
+
"grad_norm": 1.957473416803994,
|
| 2485 |
+
"learning_rate": 8.412445750974385e-08,
|
| 2486 |
+
"loss": 0.3461,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9500805152979066,
|
| 2491 |
+
"grad_norm": 2.1168590825637343,
|
| 2492 |
+
"learning_rate": 7.578051044585644e-08,
|
| 2493 |
+
"loss": 0.3501,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.95276435856146,
|
| 2498 |
+
"grad_norm": 2.052308145807329,
|
| 2499 |
+
"learning_rate": 6.786910305652373e-08,
|
| 2500 |
+
"loss": 0.3485,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9554482018250134,
|
| 2505 |
+
"grad_norm": 2.255753145623415,
|
| 2506 |
+
"learning_rate": 6.0390930273721e-08,
|
| 2507 |
+
"loss": 0.3476,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9581320450885669,
|
| 2512 |
+
"grad_norm": 2.062015912171957,
|
| 2513 |
+
"learning_rate": 5.334664897442721e-08,
|
| 2514 |
+
"loss": 0.3478,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9608158883521203,
|
| 2519 |
+
"grad_norm": 2.08850888814244,
|
| 2520 |
+
"learning_rate": 4.673687792292436e-08,
|
| 2521 |
+
"loss": 0.331,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9634997316156736,
|
| 2526 |
+
"grad_norm": 1.958352874050667,
|
| 2527 |
+
"learning_rate": 4.0562197716448316e-08,
|
| 2528 |
+
"loss": 0.346,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.966183574879227,
|
| 2533 |
+
"grad_norm": 2.199826624458542,
|
| 2534 |
+
"learning_rate": 3.4823150734186184e-08,
|
| 2535 |
+
"loss": 0.3513,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9688674181427804,
|
| 2540 |
+
"grad_norm": 1.9676144059006588,
|
| 2541 |
+
"learning_rate": 2.9520241089636692e-08,
|
| 2542 |
+
"loss": 0.3408,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9715512614063339,
|
| 2547 |
+
"grad_norm": 2.0588816398308203,
|
| 2548 |
+
"learning_rate": 2.4653934586328365e-08,
|
| 2549 |
+
"loss": 0.3476,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9742351046698873,
|
| 2554 |
+
"grad_norm": 2.0271208836113956,
|
| 2555 |
+
"learning_rate": 2.022465867690282e-08,
|
| 2556 |
+
"loss": 0.342,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9769189479334407,
|
| 2561 |
+
"grad_norm": 2.069016814300102,
|
| 2562 |
+
"learning_rate": 1.6232802425568683e-08,
|
| 2563 |
+
"loss": 0.3486,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9796027911969941,
|
| 2568 |
+
"grad_norm": 2.094116862857401,
|
| 2569 |
+
"learning_rate": 1.2678716473927821e-08,
|
| 2570 |
+
"loss": 0.3502,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9822866344605475,
|
| 2575 |
+
"grad_norm": 2.11652859366789,
|
| 2576 |
+
"learning_rate": 9.562713010171088e-09,
|
| 2577 |
+
"loss": 0.3559,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.984970477724101,
|
| 2582 |
+
"grad_norm": 1.9636086634173566,
|
| 2583 |
+
"learning_rate": 6.885065741661367e-09,
|
| 2584 |
+
"loss": 0.3462,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9876543209876543,
|
| 2589 |
+
"grad_norm": 2.0953962053297115,
|
| 2590 |
+
"learning_rate": 4.646009870885593e-09,
|
| 2591 |
+
"loss": 0.3658,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9903381642512077,
|
| 2596 |
+
"grad_norm": 1.980946754920061,
|
| 2597 |
+
"learning_rate": 2.845742074801261e-09,
|
| 2598 |
+
"loss": 0.3472,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9930220075147611,
|
| 2603 |
+
"grad_norm": 2.081291822609216,
|
| 2604 |
+
"learning_rate": 1.484420487553595e-09,
|
| 2605 |
+
"loss": 0.3509,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9957058507783145,
|
| 2610 |
+
"grad_norm": 1.962894597303225,
|
| 2611 |
+
"learning_rate": 5.621646865899832e-10,
|
| 2612 |
+
"loss": 0.339,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.998389694041868,
|
| 2617 |
+
"grad_norm": 2.014333652313711,
|
| 2618 |
+
"learning_rate": 7.905568215504921e-11,
|
| 2619 |
+
"loss": 0.3415,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 1.0,
|
| 2624 |
+
"eval_runtime": 3.3909,
|
| 2625 |
+
"eval_samples_per_second": 2.949,
|
| 2626 |
+
"eval_steps_per_second": 0.885,
|
| 2627 |
+
"step": 1863
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 1.0,
|
| 2631 |
+
"step": 1863,
|
| 2632 |
+
"total_flos": 195037149265920.0,
|
| 2633 |
+
"train_loss": 0.5120102114659513,
|
| 2634 |
+
"train_runtime": 16685.4185,
|
| 2635 |
+
"train_samples_per_second": 1.786,
|
| 2636 |
+
"train_steps_per_second": 0.112
|
| 2637 |
+
}
|
| 2638 |
+
],
|
| 2639 |
+
"logging_steps": 5,
|
| 2640 |
+
"max_steps": 1863,
|
| 2641 |
+
"num_input_tokens_seen": 0,
|
| 2642 |
+
"num_train_epochs": 1,
|
| 2643 |
+
"save_steps": 100,
|
| 2644 |
+
"stateful_callbacks": {
|
| 2645 |
+
"TrainerControl": {
|
| 2646 |
+
"args": {
|
| 2647 |
+
"should_epoch_stop": false,
|
| 2648 |
+
"should_evaluate": false,
|
| 2649 |
+
"should_log": false,
|
| 2650 |
+
"should_save": true,
|
| 2651 |
+
"should_training_stop": true
|
| 2652 |
+
},
|
| 2653 |
+
"attributes": {}
|
| 2654 |
+
}
|
| 2655 |
+
},
|
| 2656 |
+
"total_flos": 195037149265920.0,
|
| 2657 |
+
"train_batch_size": 4,
|
| 2658 |
+
"trial_name": null,
|
| 2659 |
+
"trial_params": null
|
| 2660 |
+
}
|