Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep23_20-51-54_COE-CS-sv003/events.out.tfevents.1727125016.COE-CS-sv003.741860.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2653 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: feedback_p0.1_seed42_level2_syntaxmixbatch16_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_syntaxmixbatch16_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": 194513700126720.0,
|
| 4 |
+
"train_loss": 0.5110224842126967,
|
| 5 |
+
"train_runtime": 16738.0887,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.776,
|
| 8 |
+
"train_steps_per_second": 0.111
|
| 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/Sep23_20-51-54_COE-CS-sv003/events.out.tfevents.1727125016.COE-CS-sv003.741860.0
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 sha256:7decca42eff3d9f6950a9d7bef0aad74e1aca8a181c183deb97912f156dfb001
|
| 3 |
+
size 84263
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 194513700126720.0,
|
| 4 |
+
"train_loss": 0.5110224842126967,
|
| 5 |
+
"train_runtime": 16738.0887,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.776,
|
| 8 |
+
"train_steps_per_second": 0.111
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2653 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1858,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005382131324004305,
|
| 13 |
+
"grad_norm": 24.46331800159023,
|
| 14 |
+
"learning_rate": 5.376344086021506e-08,
|
| 15 |
+
"loss": 1.3638,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002691065662002153,
|
| 20 |
+
"grad_norm": 24.232442902094746,
|
| 21 |
+
"learning_rate": 2.688172043010753e-07,
|
| 22 |
+
"loss": 1.3692,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005382131324004306,
|
| 27 |
+
"grad_norm": 14.752612983186452,
|
| 28 |
+
"learning_rate": 5.376344086021506e-07,
|
| 29 |
+
"loss": 1.3038,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008073196986006458,
|
| 34 |
+
"grad_norm": 11.800751606437952,
|
| 35 |
+
"learning_rate": 8.064516129032258e-07,
|
| 36 |
+
"loss": 1.166,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010764262648008612,
|
| 41 |
+
"grad_norm": 9.554957230427643,
|
| 42 |
+
"learning_rate": 1.0752688172043011e-06,
|
| 43 |
+
"loss": 1.0338,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013455328310010764,
|
| 48 |
+
"grad_norm": 3.628600831390835,
|
| 49 |
+
"learning_rate": 1.3440860215053765e-06,
|
| 50 |
+
"loss": 0.9317,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.016146393972012917,
|
| 55 |
+
"grad_norm": 3.2255977477470297,
|
| 56 |
+
"learning_rate": 1.6129032258064516e-06,
|
| 57 |
+
"loss": 0.9012,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.01883745963401507,
|
| 62 |
+
"grad_norm": 3.0289714775630516,
|
| 63 |
+
"learning_rate": 1.881720430107527e-06,
|
| 64 |
+
"loss": 0.8827,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.021528525296017224,
|
| 69 |
+
"grad_norm": 3.1823522788295704,
|
| 70 |
+
"learning_rate": 2.1505376344086023e-06,
|
| 71 |
+
"loss": 0.8644,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.024219590958019375,
|
| 76 |
+
"grad_norm": 2.8342525430348315,
|
| 77 |
+
"learning_rate": 2.4193548387096776e-06,
|
| 78 |
+
"loss": 0.8452,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02691065662002153,
|
| 83 |
+
"grad_norm": 3.052168326380914,
|
| 84 |
+
"learning_rate": 2.688172043010753e-06,
|
| 85 |
+
"loss": 0.8347,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.029601722282023683,
|
| 90 |
+
"grad_norm": 2.8700531898378,
|
| 91 |
+
"learning_rate": 2.9569892473118283e-06,
|
| 92 |
+
"loss": 0.8117,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03229278794402583,
|
| 97 |
+
"grad_norm": 2.9876737446045385,
|
| 98 |
+
"learning_rate": 3.225806451612903e-06,
|
| 99 |
+
"loss": 0.815,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03498385360602799,
|
| 104 |
+
"grad_norm": 3.0224919131706023,
|
| 105 |
+
"learning_rate": 3.494623655913979e-06,
|
| 106 |
+
"loss": 0.7953,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03767491926803014,
|
| 111 |
+
"grad_norm": 3.0473808806620517,
|
| 112 |
+
"learning_rate": 3.763440860215054e-06,
|
| 113 |
+
"loss": 0.7859,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.040365984930032295,
|
| 118 |
+
"grad_norm": 3.096384352129974,
|
| 119 |
+
"learning_rate": 4.032258064516129e-06,
|
| 120 |
+
"loss": 0.7971,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04305705059203445,
|
| 125 |
+
"grad_norm": 3.1948521945715074,
|
| 126 |
+
"learning_rate": 4.3010752688172045e-06,
|
| 127 |
+
"loss": 0.7941,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.045748116254036596,
|
| 132 |
+
"grad_norm": 3.0704452147023473,
|
| 133 |
+
"learning_rate": 4.56989247311828e-06,
|
| 134 |
+
"loss": 0.7655,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04843918191603875,
|
| 139 |
+
"grad_norm": 3.213509560134052,
|
| 140 |
+
"learning_rate": 4.838709677419355e-06,
|
| 141 |
+
"loss": 0.7705,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.051130247578040904,
|
| 146 |
+
"grad_norm": 3.551334383153254,
|
| 147 |
+
"learning_rate": 5.1075268817204305e-06,
|
| 148 |
+
"loss": 0.7731,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05382131324004306,
|
| 153 |
+
"grad_norm": 3.3115985648508595,
|
| 154 |
+
"learning_rate": 5.376344086021506e-06,
|
| 155 |
+
"loss": 0.7421,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05651237890204521,
|
| 160 |
+
"grad_norm": 3.024863181759173,
|
| 161 |
+
"learning_rate": 5.645161290322582e-06,
|
| 162 |
+
"loss": 0.7446,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.059203444564047365,
|
| 167 |
+
"grad_norm": 3.003509243588486,
|
| 168 |
+
"learning_rate": 5.9139784946236566e-06,
|
| 169 |
+
"loss": 0.7168,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06189451022604951,
|
| 174 |
+
"grad_norm": 2.8904939122926785,
|
| 175 |
+
"learning_rate": 6.182795698924732e-06,
|
| 176 |
+
"loss": 0.7226,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06458557588805167,
|
| 181 |
+
"grad_norm": 3.0907313428716585,
|
| 182 |
+
"learning_rate": 6.451612903225806e-06,
|
| 183 |
+
"loss": 0.7287,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06727664155005382,
|
| 188 |
+
"grad_norm": 2.908190969126188,
|
| 189 |
+
"learning_rate": 6.720430107526882e-06,
|
| 190 |
+
"loss": 0.7362,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06996770721205597,
|
| 195 |
+
"grad_norm": 3.0277469501250267,
|
| 196 |
+
"learning_rate": 6.989247311827958e-06,
|
| 197 |
+
"loss": 0.7127,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07265877287405813,
|
| 202 |
+
"grad_norm": 2.9375347724298946,
|
| 203 |
+
"learning_rate": 7.258064516129033e-06,
|
| 204 |
+
"loss": 0.7057,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07534983853606028,
|
| 209 |
+
"grad_norm": 2.7968808634431097,
|
| 210 |
+
"learning_rate": 7.526881720430108e-06,
|
| 211 |
+
"loss": 0.7212,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07804090419806244,
|
| 216 |
+
"grad_norm": 3.0782635414890342,
|
| 217 |
+
"learning_rate": 7.795698924731183e-06,
|
| 218 |
+
"loss": 0.7121,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08073196986006459,
|
| 223 |
+
"grad_norm": 2.957567985639813,
|
| 224 |
+
"learning_rate": 8.064516129032258e-06,
|
| 225 |
+
"loss": 0.7285,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08342303552206674,
|
| 230 |
+
"grad_norm": 2.8558517093443045,
|
| 231 |
+
"learning_rate": 8.333333333333334e-06,
|
| 232 |
+
"loss": 0.7182,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.0861141011840689,
|
| 237 |
+
"grad_norm": 2.875223992789043,
|
| 238 |
+
"learning_rate": 8.602150537634409e-06,
|
| 239 |
+
"loss": 0.7179,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08880516684607104,
|
| 244 |
+
"grad_norm": 3.0182318773591517,
|
| 245 |
+
"learning_rate": 8.870967741935484e-06,
|
| 246 |
+
"loss": 0.702,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09149623250807319,
|
| 251 |
+
"grad_norm": 2.7804594099140014,
|
| 252 |
+
"learning_rate": 9.13978494623656e-06,
|
| 253 |
+
"loss": 0.7105,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09418729817007535,
|
| 258 |
+
"grad_norm": 2.8174567099055157,
|
| 259 |
+
"learning_rate": 9.408602150537635e-06,
|
| 260 |
+
"loss": 0.7095,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.0968783638320775,
|
| 265 |
+
"grad_norm": 2.803920967572859,
|
| 266 |
+
"learning_rate": 9.67741935483871e-06,
|
| 267 |
+
"loss": 0.7048,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09956942949407965,
|
| 272 |
+
"grad_norm": 2.7583241887420917,
|
| 273 |
+
"learning_rate": 9.946236559139786e-06,
|
| 274 |
+
"loss": 0.7161,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10226049515608181,
|
| 279 |
+
"grad_norm": 2.9304582816909286,
|
| 280 |
+
"learning_rate": 9.999858783596665e-06,
|
| 281 |
+
"loss": 0.7136,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10495156081808396,
|
| 286 |
+
"grad_norm": 2.8592073188711886,
|
| 287 |
+
"learning_rate": 9.999285105629308e-06,
|
| 288 |
+
"loss": 0.7213,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10764262648008611,
|
| 293 |
+
"grad_norm": 2.6445409308883256,
|
| 294 |
+
"learning_rate": 9.998270190666602e-06,
|
| 295 |
+
"loss": 0.71,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.11033369214208827,
|
| 300 |
+
"grad_norm": 2.811803987053168,
|
| 301 |
+
"learning_rate": 9.99681412828496e-06,
|
| 302 |
+
"loss": 0.6946,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11302475780409042,
|
| 307 |
+
"grad_norm": 2.60382159101248,
|
| 308 |
+
"learning_rate": 9.994917046996472e-06,
|
| 309 |
+
"loss": 0.6824,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11571582346609258,
|
| 314 |
+
"grad_norm": 2.7632251330599327,
|
| 315 |
+
"learning_rate": 9.99257911423757e-06,
|
| 316 |
+
"loss": 0.712,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11840688912809473,
|
| 321 |
+
"grad_norm": 2.737121694120479,
|
| 322 |
+
"learning_rate": 9.989800536354243e-06,
|
| 323 |
+
"loss": 0.705,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.12109795479009688,
|
| 328 |
+
"grad_norm": 2.7950967614991105,
|
| 329 |
+
"learning_rate": 9.986581558583824e-06,
|
| 330 |
+
"loss": 0.6785,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12378902045209902,
|
| 335 |
+
"grad_norm": 2.580895787265613,
|
| 336 |
+
"learning_rate": 9.98292246503335e-06,
|
| 337 |
+
"loss": 0.6921,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1264800861141012,
|
| 342 |
+
"grad_norm": 2.740755897744941,
|
| 343 |
+
"learning_rate": 9.978823578654486e-06,
|
| 344 |
+
"loss": 0.7008,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12917115177610333,
|
| 349 |
+
"grad_norm": 2.5740723463717514,
|
| 350 |
+
"learning_rate": 9.97428526121502e-06,
|
| 351 |
+
"loss": 0.6851,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1318622174381055,
|
| 356 |
+
"grad_norm": 2.5648119982196316,
|
| 357 |
+
"learning_rate": 9.969307913266931e-06,
|
| 358 |
+
"loss": 0.6918,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13455328310010764,
|
| 363 |
+
"grad_norm": 2.6146329921031968,
|
| 364 |
+
"learning_rate": 9.963891974111042e-06,
|
| 365 |
+
"loss": 0.686,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1372443487621098,
|
| 370 |
+
"grad_norm": 2.5693313714525225,
|
| 371 |
+
"learning_rate": 9.958037921758241e-06,
|
| 372 |
+
"loss": 0.6851,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13993541442411195,
|
| 377 |
+
"grad_norm": 2.5758611472453183,
|
| 378 |
+
"learning_rate": 9.951746272887298e-06,
|
| 379 |
+
"loss": 0.6825,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1426264800861141,
|
| 384 |
+
"grad_norm": 2.6882089017132866,
|
| 385 |
+
"learning_rate": 9.945017582799256e-06,
|
| 386 |
+
"loss": 0.6857,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14531754574811626,
|
| 391 |
+
"grad_norm": 2.471441857805247,
|
| 392 |
+
"learning_rate": 9.937852445368427e-06,
|
| 393 |
+
"loss": 0.7087,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1480086114101184,
|
| 398 |
+
"grad_norm": 2.499362888582163,
|
| 399 |
+
"learning_rate": 9.930251492989972e-06,
|
| 400 |
+
"loss": 0.6834,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.15069967707212056,
|
| 405 |
+
"grad_norm": 2.4532511507065684,
|
| 406 |
+
"learning_rate": 9.922215396524089e-06,
|
| 407 |
+
"loss": 0.6841,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1533907427341227,
|
| 412 |
+
"grad_norm": 2.476586585638722,
|
| 413 |
+
"learning_rate": 9.913744865236798e-06,
|
| 414 |
+
"loss": 0.6738,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.15608180839612487,
|
| 419 |
+
"grad_norm": 2.4741617401097593,
|
| 420 |
+
"learning_rate": 9.904840646737346e-06,
|
| 421 |
+
"loss": 0.6977,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.158772874058127,
|
| 426 |
+
"grad_norm": 2.4646734777776405,
|
| 427 |
+
"learning_rate": 9.895503526912224e-06,
|
| 428 |
+
"loss": 0.6664,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.16146393972012918,
|
| 433 |
+
"grad_norm": 2.541571671880384,
|
| 434 |
+
"learning_rate": 9.885734329855798e-06,
|
| 435 |
+
"loss": 0.6578,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.16415500538213132,
|
| 440 |
+
"grad_norm": 2.512328994032953,
|
| 441 |
+
"learning_rate": 9.875533917797579e-06,
|
| 442 |
+
"loss": 0.6564,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.1668460710441335,
|
| 447 |
+
"grad_norm": 2.608219630629651,
|
| 448 |
+
"learning_rate": 9.864903191026125e-06,
|
| 449 |
+
"loss": 0.6676,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16953713670613563,
|
| 454 |
+
"grad_norm": 2.5127310208806226,
|
| 455 |
+
"learning_rate": 9.853843087809574e-06,
|
| 456 |
+
"loss": 0.675,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1722282023681378,
|
| 461 |
+
"grad_norm": 2.954995883472683,
|
| 462 |
+
"learning_rate": 9.842354584312841e-06,
|
| 463 |
+
"loss": 0.6711,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17491926803013993,
|
| 468 |
+
"grad_norm": 2.4877912144296723,
|
| 469 |
+
"learning_rate": 9.830438694511454e-06,
|
| 470 |
+
"loss": 0.6532,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17761033369214208,
|
| 475 |
+
"grad_norm": 2.4109884507809998,
|
| 476 |
+
"learning_rate": 9.818096470102067e-06,
|
| 477 |
+
"loss": 0.6742,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.18030139935414424,
|
| 482 |
+
"grad_norm": 2.7208831806314633,
|
| 483 |
+
"learning_rate": 9.805329000409634e-06,
|
| 484 |
+
"loss": 0.6626,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.18299246501614638,
|
| 489 |
+
"grad_norm": 2.70446788462542,
|
| 490 |
+
"learning_rate": 9.792137412291265e-06,
|
| 491 |
+
"loss": 0.6578,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18568353067814855,
|
| 496 |
+
"grad_norm": 2.554719643429019,
|
| 497 |
+
"learning_rate": 9.778522870036768e-06,
|
| 498 |
+
"loss": 0.6386,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1883745963401507,
|
| 503 |
+
"grad_norm": 2.548018478530073,
|
| 504 |
+
"learning_rate": 9.764486575265893e-06,
|
| 505 |
+
"loss": 0.653,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.19106566200215286,
|
| 510 |
+
"grad_norm": 2.720938193286558,
|
| 511 |
+
"learning_rate": 9.750029766822277e-06,
|
| 512 |
+
"loss": 0.6579,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.193756727664155,
|
| 517 |
+
"grad_norm": 3.2604834173029267,
|
| 518 |
+
"learning_rate": 9.735153720664096e-06,
|
| 519 |
+
"loss": 0.6357,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19644779332615717,
|
| 524 |
+
"grad_norm": 2.854496420392015,
|
| 525 |
+
"learning_rate": 9.719859749751462e-06,
|
| 526 |
+
"loss": 0.643,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1991388589881593,
|
| 531 |
+
"grad_norm": 2.5148923925605295,
|
| 532 |
+
"learning_rate": 9.704149203930522e-06,
|
| 533 |
+
"loss": 0.6314,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.20182992465016147,
|
| 538 |
+
"grad_norm": 2.6156775549028097,
|
| 539 |
+
"learning_rate": 9.688023469814345e-06,
|
| 540 |
+
"loss": 0.6291,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.20452099031216361,
|
| 545 |
+
"grad_norm": 2.4436480149075903,
|
| 546 |
+
"learning_rate": 9.671483970660519e-06,
|
| 547 |
+
"loss": 0.6391,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20721205597416578,
|
| 552 |
+
"grad_norm": 2.6176248575730683,
|
| 553 |
+
"learning_rate": 9.654532166245543e-06,
|
| 554 |
+
"loss": 0.6451,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20990312163616792,
|
| 559 |
+
"grad_norm": 2.6284920127628704,
|
| 560 |
+
"learning_rate": 9.637169552735985e-06,
|
| 561 |
+
"loss": 0.6457,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.21259418729817006,
|
| 566 |
+
"grad_norm": 2.5079900288395676,
|
| 567 |
+
"learning_rate": 9.619397662556434e-06,
|
| 568 |
+
"loss": 0.6363,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.21528525296017223,
|
| 573 |
+
"grad_norm": 2.510239111931181,
|
| 574 |
+
"learning_rate": 9.601218064254245e-06,
|
| 575 |
+
"loss": 0.636,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.21797631862217437,
|
| 580 |
+
"grad_norm": 2.6420717433236294,
|
| 581 |
+
"learning_rate": 9.582632362361098e-06,
|
| 582 |
+
"loss": 0.638,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.22066738428417654,
|
| 587 |
+
"grad_norm": 2.4228761926249343,
|
| 588 |
+
"learning_rate": 9.563642197251382e-06,
|
| 589 |
+
"loss": 0.6373,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.22335844994617868,
|
| 594 |
+
"grad_norm": 2.482460770882953,
|
| 595 |
+
"learning_rate": 9.54424924499742e-06,
|
| 596 |
+
"loss": 0.6138,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22604951560818085,
|
| 601 |
+
"grad_norm": 2.4554209699636904,
|
| 602 |
+
"learning_rate": 9.524455217221537e-06,
|
| 603 |
+
"loss": 0.6297,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22874058127018299,
|
| 608 |
+
"grad_norm": 2.2267259427452686,
|
| 609 |
+
"learning_rate": 9.504261860944984e-06,
|
| 610 |
+
"loss": 0.6345,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.23143164693218515,
|
| 615 |
+
"grad_norm": 2.5062964865846467,
|
| 616 |
+
"learning_rate": 9.48367095843376e-06,
|
| 617 |
+
"loss": 0.6271,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2341227125941873,
|
| 622 |
+
"grad_norm": 2.5483397828557353,
|
| 623 |
+
"learning_rate": 9.462684327041298e-06,
|
| 624 |
+
"loss": 0.6137,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23681377825618946,
|
| 629 |
+
"grad_norm": 6.340986217773637,
|
| 630 |
+
"learning_rate": 9.441303819048073e-06,
|
| 631 |
+
"loss": 0.6236,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2395048439181916,
|
| 636 |
+
"grad_norm": 3.6533277008246787,
|
| 637 |
+
"learning_rate": 9.41953132149811e-06,
|
| 638 |
+
"loss": 0.6201,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.24219590958019377,
|
| 643 |
+
"grad_norm": 2.539798703180873,
|
| 644 |
+
"learning_rate": 9.397368756032445e-06,
|
| 645 |
+
"loss": 0.623,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2448869752421959,
|
| 650 |
+
"grad_norm": 2.569421112776594,
|
| 651 |
+
"learning_rate": 9.374818078719515e-06,
|
| 652 |
+
"loss": 0.6129,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24757804090419805,
|
| 657 |
+
"grad_norm": 2.3715746149053953,
|
| 658 |
+
"learning_rate": 9.351881279882512e-06,
|
| 659 |
+
"loss": 0.6268,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2502691065662002,
|
| 664 |
+
"grad_norm": 2.572579471719146,
|
| 665 |
+
"learning_rate": 9.328560383923724e-06,
|
| 666 |
+
"loss": 0.6161,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2529601722282024,
|
| 671 |
+
"grad_norm": 2.4902601550757737,
|
| 672 |
+
"learning_rate": 9.304857449145858e-06,
|
| 673 |
+
"loss": 0.6244,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2556512378902045,
|
| 678 |
+
"grad_norm": 2.4501823079061658,
|
| 679 |
+
"learning_rate": 9.280774567570372e-06,
|
| 680 |
+
"loss": 0.6287,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.25834230355220666,
|
| 685 |
+
"grad_norm": 2.5682717978671126,
|
| 686 |
+
"learning_rate": 9.256313864752838e-06,
|
| 687 |
+
"loss": 0.604,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.26103336921420883,
|
| 692 |
+
"grad_norm": 2.404883352610733,
|
| 693 |
+
"learning_rate": 9.231477499595333e-06,
|
| 694 |
+
"loss": 0.6138,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.263724434876211,
|
| 699 |
+
"grad_norm": 2.562822568126396,
|
| 700 |
+
"learning_rate": 9.206267664155906e-06,
|
| 701 |
+
"loss": 0.6054,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2664155005382131,
|
| 706 |
+
"grad_norm": 2.321411236526974,
|
| 707 |
+
"learning_rate": 9.180686583455097e-06,
|
| 708 |
+
"loss": 0.5948,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2691065662002153,
|
| 713 |
+
"grad_norm": 2.475132775898768,
|
| 714 |
+
"learning_rate": 9.154736515279557e-06,
|
| 715 |
+
"loss": 0.5905,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.27179763186221745,
|
| 720 |
+
"grad_norm": 2.3355547839115354,
|
| 721 |
+
"learning_rate": 9.12841974998278e-06,
|
| 722 |
+
"loss": 0.5813,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2744886975242196,
|
| 727 |
+
"grad_norm": 2.742626737076485,
|
| 728 |
+
"learning_rate": 9.101738610282956e-06,
|
| 729 |
+
"loss": 0.6136,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.27717976318622173,
|
| 734 |
+
"grad_norm": 2.52218472070317,
|
| 735 |
+
"learning_rate": 9.074695451057966e-06,
|
| 736 |
+
"loss": 0.6002,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.2798708288482239,
|
| 741 |
+
"grad_norm": 2.364034819036867,
|
| 742 |
+
"learning_rate": 9.047292659137542e-06,
|
| 743 |
+
"loss": 0.6055,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.28256189451022606,
|
| 748 |
+
"grad_norm": 2.5115978707447524,
|
| 749 |
+
"learning_rate": 9.019532653092597e-06,
|
| 750 |
+
"loss": 0.5978,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2852529601722282,
|
| 755 |
+
"grad_norm": 2.4450725301262457,
|
| 756 |
+
"learning_rate": 8.99141788302178e-06,
|
| 757 |
+
"loss": 0.5748,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.28794402583423034,
|
| 762 |
+
"grad_norm": 2.5635891607971764,
|
| 763 |
+
"learning_rate": 8.962950830335213e-06,
|
| 764 |
+
"loss": 0.6019,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2906350914962325,
|
| 769 |
+
"grad_norm": 2.6838630282585316,
|
| 770 |
+
"learning_rate": 8.93413400753549e-06,
|
| 771 |
+
"loss": 0.6119,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2933261571582347,
|
| 776 |
+
"grad_norm": 2.559914467821027,
|
| 777 |
+
"learning_rate": 8.90496995799592e-06,
|
| 778 |
+
"loss": 0.5934,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.2960172228202368,
|
| 783 |
+
"grad_norm": 2.3172236979073633,
|
| 784 |
+
"learning_rate": 8.875461255736055e-06,
|
| 785 |
+
"loss": 0.5923,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.29870828848223896,
|
| 790 |
+
"grad_norm": 2.613084787164224,
|
| 791 |
+
"learning_rate": 8.845610505194495e-06,
|
| 792 |
+
"loss": 0.5881,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3013993541442411,
|
| 797 |
+
"grad_norm": 2.527915276378667,
|
| 798 |
+
"learning_rate": 8.815420340999034e-06,
|
| 799 |
+
"loss": 0.5877,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3040904198062433,
|
| 804 |
+
"grad_norm": 2.566801330647483,
|
| 805 |
+
"learning_rate": 8.784893427734117e-06,
|
| 806 |
+
"loss": 0.5742,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3067814854682454,
|
| 811 |
+
"grad_norm": 2.363950845115375,
|
| 812 |
+
"learning_rate": 8.754032459705672e-06,
|
| 813 |
+
"loss": 0.5828,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3094725511302476,
|
| 818 |
+
"grad_norm": 2.2951780141802307,
|
| 819 |
+
"learning_rate": 8.722840160703304e-06,
|
| 820 |
+
"loss": 0.5825,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.31216361679224974,
|
| 825 |
+
"grad_norm": 2.53863020294738,
|
| 826 |
+
"learning_rate": 8.691319283759896e-06,
|
| 827 |
+
"loss": 0.5751,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3148546824542519,
|
| 832 |
+
"grad_norm": 2.357318941568188,
|
| 833 |
+
"learning_rate": 8.659472610908628e-06,
|
| 834 |
+
"loss": 0.5897,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.317545748116254,
|
| 839 |
+
"grad_norm": 2.318670881116376,
|
| 840 |
+
"learning_rate": 8.627302952937431e-06,
|
| 841 |
+
"loss": 0.5649,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3202368137782562,
|
| 846 |
+
"grad_norm": 2.2618602596217388,
|
| 847 |
+
"learning_rate": 8.594813149140908e-06,
|
| 848 |
+
"loss": 0.577,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.32292787944025836,
|
| 853 |
+
"grad_norm": 2.39290329100585,
|
| 854 |
+
"learning_rate": 8.56200606706974e-06,
|
| 855 |
+
"loss": 0.5543,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.32561894510226047,
|
| 860 |
+
"grad_norm": 2.470727799467087,
|
| 861 |
+
"learning_rate": 8.528884602277593e-06,
|
| 862 |
+
"loss": 0.5823,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.32831001076426264,
|
| 867 |
+
"grad_norm": 2.4343415189492403,
|
| 868 |
+
"learning_rate": 8.495451678065563e-06,
|
| 869 |
+
"loss": 0.583,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3310010764262648,
|
| 874 |
+
"grad_norm": 2.329527413930226,
|
| 875 |
+
"learning_rate": 8.461710245224149e-06,
|
| 876 |
+
"loss": 0.5596,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.333692142088267,
|
| 881 |
+
"grad_norm": 2.710197180873677,
|
| 882 |
+
"learning_rate": 8.42766328177284e-06,
|
| 883 |
+
"loss": 0.5688,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3363832077502691,
|
| 888 |
+
"grad_norm": 2.3295706886393908,
|
| 889 |
+
"learning_rate": 8.393313792697251e-06,
|
| 890 |
+
"loss": 0.5606,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.33907427341227125,
|
| 895 |
+
"grad_norm": 2.7411314085447893,
|
| 896 |
+
"learning_rate": 8.358664809683926e-06,
|
| 897 |
+
"loss": 0.5679,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3417653390742734,
|
| 902 |
+
"grad_norm": 2.354293086420231,
|
| 903 |
+
"learning_rate": 8.323719390852735e-06,
|
| 904 |
+
"loss": 0.5737,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3444564047362756,
|
| 909 |
+
"grad_norm": 2.5084140873961083,
|
| 910 |
+
"learning_rate": 8.288480620486991e-06,
|
| 911 |
+
"loss": 0.5479,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3471474703982777,
|
| 916 |
+
"grad_norm": 2.4064859741841427,
|
| 917 |
+
"learning_rate": 8.252951608761217e-06,
|
| 918 |
+
"loss": 0.5574,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.34983853606027987,
|
| 923 |
+
"grad_norm": 2.4777423126966185,
|
| 924 |
+
"learning_rate": 8.217135491466636e-06,
|
| 925 |
+
"loss": 0.5666,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.35252960172228204,
|
| 930 |
+
"grad_norm": 2.3905287579939416,
|
| 931 |
+
"learning_rate": 8.181035429734423e-06,
|
| 932 |
+
"loss": 0.5629,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.35522066738428415,
|
| 937 |
+
"grad_norm": 2.327857140876721,
|
| 938 |
+
"learning_rate": 8.144654609756685e-06,
|
| 939 |
+
"loss": 0.5372,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3579117330462863,
|
| 944 |
+
"grad_norm": 2.290053555489138,
|
| 945 |
+
"learning_rate": 8.10799624250527e-06,
|
| 946 |
+
"loss": 0.5646,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3606027987082885,
|
| 951 |
+
"grad_norm": 2.297702496538287,
|
| 952 |
+
"learning_rate": 8.071063563448341e-06,
|
| 953 |
+
"loss": 0.553,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.36329386437029065,
|
| 958 |
+
"grad_norm": 2.302044112251165,
|
| 959 |
+
"learning_rate": 8.03385983226483e-06,
|
| 960 |
+
"loss": 0.5381,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.36598493003229277,
|
| 965 |
+
"grad_norm": 2.508769537248855,
|
| 966 |
+
"learning_rate": 7.996388332556735e-06,
|
| 967 |
+
"loss": 0.5433,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.36867599569429493,
|
| 972 |
+
"grad_norm": 2.4009515047198193,
|
| 973 |
+
"learning_rate": 7.958652371559313e-06,
|
| 974 |
+
"loss": 0.5524,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3713670613562971,
|
| 979 |
+
"grad_norm": 2.3402259477581198,
|
| 980 |
+
"learning_rate": 7.920655279849173e-06,
|
| 981 |
+
"loss": 0.5609,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.37405812701829927,
|
| 986 |
+
"grad_norm": 2.71945405647194,
|
| 987 |
+
"learning_rate": 7.882400411050328e-06,
|
| 988 |
+
"loss": 0.5414,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3767491926803014,
|
| 993 |
+
"grad_norm": 2.325373282991122,
|
| 994 |
+
"learning_rate": 7.843891141538201e-06,
|
| 995 |
+
"loss": 0.5352,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.37944025834230355,
|
| 1000 |
+
"grad_norm": 2.243347153689357,
|
| 1001 |
+
"learning_rate": 7.80513087014163e-06,
|
| 1002 |
+
"loss": 0.5503,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3821313240043057,
|
| 1007 |
+
"grad_norm": 2.2504693819243506,
|
| 1008 |
+
"learning_rate": 7.766123017842877e-06,
|
| 1009 |
+
"loss": 0.5408,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3848223896663079,
|
| 1014 |
+
"grad_norm": 2.24546531050267,
|
| 1015 |
+
"learning_rate": 7.726871027475709e-06,
|
| 1016 |
+
"loss": 0.5384,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.38751345532831,
|
| 1021 |
+
"grad_norm": 2.452293080065367,
|
| 1022 |
+
"learning_rate": 7.687378363421512e-06,
|
| 1023 |
+
"loss": 0.5435,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.39020452099031216,
|
| 1028 |
+
"grad_norm": 2.4702348319235625,
|
| 1029 |
+
"learning_rate": 7.647648511303545e-06,
|
| 1030 |
+
"loss": 0.5366,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.39289558665231433,
|
| 1035 |
+
"grad_norm": 2.6594839428477646,
|
| 1036 |
+
"learning_rate": 7.607684977679284e-06,
|
| 1037 |
+
"loss": 0.5311,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.39558665231431644,
|
| 1042 |
+
"grad_norm": 2.3853401333086097,
|
| 1043 |
+
"learning_rate": 7.567491289730944e-06,
|
| 1044 |
+
"loss": 0.5347,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3982777179763186,
|
| 1049 |
+
"grad_norm": 2.305805010232876,
|
| 1050 |
+
"learning_rate": 7.52707099495416e-06,
|
| 1051 |
+
"loss": 0.5198,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.4009687836383208,
|
| 1056 |
+
"grad_norm": 2.32709407366988,
|
| 1057 |
+
"learning_rate": 7.4864276608448925e-06,
|
| 1058 |
+
"loss": 0.5162,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.40365984930032295,
|
| 1063 |
+
"grad_norm": 2.2558330926103993,
|
| 1064 |
+
"learning_rate": 7.44556487458456e-06,
|
| 1065 |
+
"loss": 0.5157,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.40635091496232506,
|
| 1070 |
+
"grad_norm": 2.358890714175164,
|
| 1071 |
+
"learning_rate": 7.404486242723428e-06,
|
| 1072 |
+
"loss": 0.5223,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.40904198062432723,
|
| 1077 |
+
"grad_norm": 2.4688588460570036,
|
| 1078 |
+
"learning_rate": 7.363195390862298e-06,
|
| 1079 |
+
"loss": 0.5306,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4117330462863294,
|
| 1084 |
+
"grad_norm": 2.342696497908346,
|
| 1085 |
+
"learning_rate": 7.321695963332516e-06,
|
| 1086 |
+
"loss": 0.5331,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.41442411194833156,
|
| 1091 |
+
"grad_norm": 2.5064305240289144,
|
| 1092 |
+
"learning_rate": 7.279991622874319e-06,
|
| 1093 |
+
"loss": 0.5397,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4171151776103337,
|
| 1098 |
+
"grad_norm": 2.3898169886899296,
|
| 1099 |
+
"learning_rate": 7.238086050313563e-06,
|
| 1100 |
+
"loss": 0.5197,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.41980624327233584,
|
| 1105 |
+
"grad_norm": 2.429842329992259,
|
| 1106 |
+
"learning_rate": 7.195982944236853e-06,
|
| 1107 |
+
"loss": 0.516,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.422497308934338,
|
| 1112 |
+
"grad_norm": 2.3477023797859813,
|
| 1113 |
+
"learning_rate": 7.1536860206651025e-06,
|
| 1114 |
+
"loss": 0.5215,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4251883745963401,
|
| 1119 |
+
"grad_norm": 2.3798189210672587,
|
| 1120 |
+
"learning_rate": 7.1111990127255684e-06,
|
| 1121 |
+
"loss": 0.5033,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4278794402583423,
|
| 1126 |
+
"grad_norm": 2.400485115987879,
|
| 1127 |
+
"learning_rate": 7.068525670322349e-06,
|
| 1128 |
+
"loss": 0.5301,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.43057050592034446,
|
| 1133 |
+
"grad_norm": 2.830405312883021,
|
| 1134 |
+
"learning_rate": 7.025669759805431e-06,
|
| 1135 |
+
"loss": 0.5226,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4332615715823466,
|
| 1140 |
+
"grad_norm": 2.267489328220382,
|
| 1141 |
+
"learning_rate": 6.982635063638265e-06,
|
| 1142 |
+
"loss": 0.5065,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.43595263724434874,
|
| 1147 |
+
"grad_norm": 2.452358924939666,
|
| 1148 |
+
"learning_rate": 6.939425380063924e-06,
|
| 1149 |
+
"loss": 0.5037,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4386437029063509,
|
| 1154 |
+
"grad_norm": 2.4030600337466588,
|
| 1155 |
+
"learning_rate": 6.896044522769879e-06,
|
| 1156 |
+
"loss": 0.5181,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4413347685683531,
|
| 1161 |
+
"grad_norm": 2.385954958028487,
|
| 1162 |
+
"learning_rate": 6.852496320551387e-06,
|
| 1163 |
+
"loss": 0.513,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.44402583423035524,
|
| 1168 |
+
"grad_norm": 2.3598226733949064,
|
| 1169 |
+
"learning_rate": 6.808784616973581e-06,
|
| 1170 |
+
"loss": 0.5126,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.44671689989235736,
|
| 1175 |
+
"grad_norm": 2.3565314879299053,
|
| 1176 |
+
"learning_rate": 6.76491327003222e-06,
|
| 1177 |
+
"loss": 0.5029,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4494079655543595,
|
| 1182 |
+
"grad_norm": 2.5160272506828756,
|
| 1183 |
+
"learning_rate": 6.720886151813194e-06,
|
| 1184 |
+
"loss": 0.4934,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.4520990312163617,
|
| 1189 |
+
"grad_norm": 2.386495536841602,
|
| 1190 |
+
"learning_rate": 6.676707148150763e-06,
|
| 1191 |
+
"loss": 0.5032,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.45479009687836386,
|
| 1196 |
+
"grad_norm": 2.436343528178474,
|
| 1197 |
+
"learning_rate": 6.632380158284607e-06,
|
| 1198 |
+
"loss": 0.4946,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.45748116254036597,
|
| 1203 |
+
"grad_norm": 2.4789859597196426,
|
| 1204 |
+
"learning_rate": 6.587909094515663e-06,
|
| 1205 |
+
"loss": 0.5066,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.46017222820236814,
|
| 1210 |
+
"grad_norm": 2.417011825533269,
|
| 1211 |
+
"learning_rate": 6.5432978818608395e-06,
|
| 1212 |
+
"loss": 0.5112,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4628632938643703,
|
| 1217 |
+
"grad_norm": 2.3434112329726275,
|
| 1218 |
+
"learning_rate": 6.498550457706584e-06,
|
| 1219 |
+
"loss": 0.5045,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4655543595263724,
|
| 1224 |
+
"grad_norm": 2.517810982631902,
|
| 1225 |
+
"learning_rate": 6.453670771461377e-06,
|
| 1226 |
+
"loss": 0.4905,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4682454251883746,
|
| 1231 |
+
"grad_norm": 2.3345178470528385,
|
| 1232 |
+
"learning_rate": 6.408662784207149e-06,
|
| 1233 |
+
"loss": 0.5141,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.47093649085037675,
|
| 1238 |
+
"grad_norm": 2.39523867329611,
|
| 1239 |
+
"learning_rate": 6.363530468349686e-06,
|
| 1240 |
+
"loss": 0.4879,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4736275565123789,
|
| 1245 |
+
"grad_norm": 2.308366752180236,
|
| 1246 |
+
"learning_rate": 6.318277807268013e-06,
|
| 1247 |
+
"loss": 0.4956,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.47631862217438103,
|
| 1252 |
+
"grad_norm": 2.2009448335762127,
|
| 1253 |
+
"learning_rate": 6.27290879496283e-06,
|
| 1254 |
+
"loss": 0.4946,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4790096878363832,
|
| 1259 |
+
"grad_norm": 2.506472513608515,
|
| 1260 |
+
"learning_rate": 6.227427435703997e-06,
|
| 1261 |
+
"loss": 0.4747,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.48170075349838537,
|
| 1266 |
+
"grad_norm": 2.3312661677886157,
|
| 1267 |
+
"learning_rate": 6.181837743677118e-06,
|
| 1268 |
+
"loss": 0.4781,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.48439181916038754,
|
| 1273 |
+
"grad_norm": 2.415462314886686,
|
| 1274 |
+
"learning_rate": 6.136143742629252e-06,
|
| 1275 |
+
"loss": 0.4846,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.48708288482238965,
|
| 1280 |
+
"grad_norm": 2.5220206493746997,
|
| 1281 |
+
"learning_rate": 6.09034946551377e-06,
|
| 1282 |
+
"loss": 0.4891,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4897739504843918,
|
| 1287 |
+
"grad_norm": 2.4607305732947853,
|
| 1288 |
+
"learning_rate": 6.044458954134411e-06,
|
| 1289 |
+
"loss": 0.5057,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.492465016146394,
|
| 1294 |
+
"grad_norm": 2.306884124252388,
|
| 1295 |
+
"learning_rate": 5.998476258788555e-06,
|
| 1296 |
+
"loss": 0.477,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4951560818083961,
|
| 1301 |
+
"grad_norm": 2.385382261424978,
|
| 1302 |
+
"learning_rate": 5.952405437909738e-06,
|
| 1303 |
+
"loss": 0.4812,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.49784714747039827,
|
| 1308 |
+
"grad_norm": 2.2862418710775847,
|
| 1309 |
+
"learning_rate": 5.90625055770946e-06,
|
| 1310 |
+
"loss": 0.4784,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5005382131324004,
|
| 1315 |
+
"grad_norm": 2.3522872723971093,
|
| 1316 |
+
"learning_rate": 5.860015691818292e-06,
|
| 1317 |
+
"loss": 0.4849,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5032292787944026,
|
| 1322 |
+
"grad_norm": 2.257405887575007,
|
| 1323 |
+
"learning_rate": 5.813704920926352e-06,
|
| 1324 |
+
"loss": 0.4799,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5059203444564048,
|
| 1329 |
+
"grad_norm": 2.5409180122003225,
|
| 1330 |
+
"learning_rate": 5.767322332423128e-06,
|
| 1331 |
+
"loss": 0.4753,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5086114101184069,
|
| 1336 |
+
"grad_norm": 2.2912568364325137,
|
| 1337 |
+
"learning_rate": 5.720872020036734e-06,
|
| 1338 |
+
"loss": 0.4836,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.511302475780409,
|
| 1343 |
+
"grad_norm": 2.3404279089465256,
|
| 1344 |
+
"learning_rate": 5.674358083472598e-06,
|
| 1345 |
+
"loss": 0.4709,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5139935414424112,
|
| 1350 |
+
"grad_norm": 2.3138319616276504,
|
| 1351 |
+
"learning_rate": 5.6277846280516125e-06,
|
| 1352 |
+
"loss": 0.4775,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5166846071044133,
|
| 1357 |
+
"grad_norm": 2.319930769748735,
|
| 1358 |
+
"learning_rate": 5.581155764347812e-06,
|
| 1359 |
+
"loss": 0.4737,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5193756727664155,
|
| 1364 |
+
"grad_norm": 2.2631450594519573,
|
| 1365 |
+
"learning_rate": 5.534475607825566e-06,
|
| 1366 |
+
"loss": 0.4681,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5220667384284177,
|
| 1371 |
+
"grad_norm": 2.2993752726651877,
|
| 1372 |
+
"learning_rate": 5.487748278476342e-06,
|
| 1373 |
+
"loss": 0.4744,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5247578040904198,
|
| 1378 |
+
"grad_norm": 2.6084393355683853,
|
| 1379 |
+
"learning_rate": 5.440977900455093e-06,
|
| 1380 |
+
"loss": 0.474,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.527448869752422,
|
| 1385 |
+
"grad_norm": 2.2812262068529767,
|
| 1386 |
+
"learning_rate": 5.39416860171624e-06,
|
| 1387 |
+
"loss": 0.4603,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5301399354144241,
|
| 1392 |
+
"grad_norm": 2.366439732267079,
|
| 1393 |
+
"learning_rate": 5.347324513649352e-06,
|
| 1394 |
+
"loss": 0.4554,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5328310010764262,
|
| 1399 |
+
"grad_norm": 2.282327173713366,
|
| 1400 |
+
"learning_rate": 5.300449770714502e-06,
|
| 1401 |
+
"loss": 0.4484,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5355220667384284,
|
| 1406 |
+
"grad_norm": 2.388092616206348,
|
| 1407 |
+
"learning_rate": 5.253548510077366e-06,
|
| 1408 |
+
"loss": 0.4565,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5382131324004306,
|
| 1413 |
+
"grad_norm": 2.347808334356873,
|
| 1414 |
+
"learning_rate": 5.206624871244066e-06,
|
| 1415 |
+
"loss": 0.4581,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5409041980624327,
|
| 1420 |
+
"grad_norm": 2.2379521568960206,
|
| 1421 |
+
"learning_rate": 5.159682995695833e-06,
|
| 1422 |
+
"loss": 0.4477,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5435952637244349,
|
| 1427 |
+
"grad_norm": 2.2336687017970296,
|
| 1428 |
+
"learning_rate": 5.112727026523461e-06,
|
| 1429 |
+
"loss": 0.4559,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5462863293864371,
|
| 1434 |
+
"grad_norm": 2.3526152961096223,
|
| 1435 |
+
"learning_rate": 5.065761108061658e-06,
|
| 1436 |
+
"loss": 0.4517,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5489773950484392,
|
| 1441 |
+
"grad_norm": 2.3430650960275305,
|
| 1442 |
+
"learning_rate": 5.018789385523245e-06,
|
| 1443 |
+
"loss": 0.4651,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5516684607104413,
|
| 1448 |
+
"grad_norm": 2.238164378282626,
|
| 1449 |
+
"learning_rate": 4.971816004633323e-06,
|
| 1450 |
+
"loss": 0.4461,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5543595263724435,
|
| 1455 |
+
"grad_norm": 2.357723214292236,
|
| 1456 |
+
"learning_rate": 4.924845111263349e-06,
|
| 1457 |
+
"loss": 0.4475,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5570505920344456,
|
| 1462 |
+
"grad_norm": 2.416110748499095,
|
| 1463 |
+
"learning_rate": 4.877880851065238e-06,
|
| 1464 |
+
"loss": 0.4621,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5597416576964478,
|
| 1469 |
+
"grad_norm": 2.446048600948981,
|
| 1470 |
+
"learning_rate": 4.830927369105457e-06,
|
| 1471 |
+
"loss": 0.4585,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.56243272335845,
|
| 1476 |
+
"grad_norm": 2.381977000693451,
|
| 1477 |
+
"learning_rate": 4.783988809499187e-06,
|
| 1478 |
+
"loss": 0.4544,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5651237890204521,
|
| 1483 |
+
"grad_norm": 2.633241388056459,
|
| 1484 |
+
"learning_rate": 4.737069315044562e-06,
|
| 1485 |
+
"loss": 0.44,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5678148546824543,
|
| 1490 |
+
"grad_norm": 2.2816953591294937,
|
| 1491 |
+
"learning_rate": 4.690173026857028e-06,
|
| 1492 |
+
"loss": 0.4501,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5705059203444564,
|
| 1497 |
+
"grad_norm": 2.446396815822862,
|
| 1498 |
+
"learning_rate": 4.643304084003839e-06,
|
| 1499 |
+
"loss": 0.4506,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5731969860064585,
|
| 1504 |
+
"grad_norm": 2.184428941729192,
|
| 1505 |
+
"learning_rate": 4.596466623138756e-06,
|
| 1506 |
+
"loss": 0.4308,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5758880516684607,
|
| 1511 |
+
"grad_norm": 2.3003382421283844,
|
| 1512 |
+
"learning_rate": 4.549664778136933e-06,
|
| 1513 |
+
"loss": 0.4416,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5785791173304629,
|
| 1518 |
+
"grad_norm": 2.2767749774728783,
|
| 1519 |
+
"learning_rate": 4.502902679730074e-06,
|
| 1520 |
+
"loss": 0.4315,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.581270182992465,
|
| 1525 |
+
"grad_norm": 2.3411661207469674,
|
| 1526 |
+
"learning_rate": 4.456184455141843e-06,
|
| 1527 |
+
"loss": 0.447,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5839612486544672,
|
| 1532 |
+
"grad_norm": 2.381303755147278,
|
| 1533 |
+
"learning_rate": 4.4095142277236015e-06,
|
| 1534 |
+
"loss": 0.4397,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5866523143164694,
|
| 1539 |
+
"grad_norm": 2.332530420483932,
|
| 1540 |
+
"learning_rate": 4.362896116590475e-06,
|
| 1541 |
+
"loss": 0.4392,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5893433799784715,
|
| 1546 |
+
"grad_norm": 2.3579190260561704,
|
| 1547 |
+
"learning_rate": 4.316334236257818e-06,
|
| 1548 |
+
"loss": 0.4328,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5920344456404736,
|
| 1553 |
+
"grad_norm": 2.2783721282271685,
|
| 1554 |
+
"learning_rate": 4.269832696278038e-06,
|
| 1555 |
+
"loss": 0.4336,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5947255113024758,
|
| 1560 |
+
"grad_norm": 2.300306639814476,
|
| 1561 |
+
"learning_rate": 4.223395600877912e-06,
|
| 1562 |
+
"loss": 0.4242,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5974165769644779,
|
| 1567 |
+
"grad_norm": 2.351752301438986,
|
| 1568 |
+
"learning_rate": 4.17702704859633e-06,
|
| 1569 |
+
"loss": 0.4442,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6001076426264801,
|
| 1574 |
+
"grad_norm": 2.4962239067091847,
|
| 1575 |
+
"learning_rate": 4.130731131922574e-06,
|
| 1576 |
+
"loss": 0.4378,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6027987082884823,
|
| 1581 |
+
"grad_norm": 2.361239331500982,
|
| 1582 |
+
"learning_rate": 4.0845119369350995e-06,
|
| 1583 |
+
"loss": 0.4262,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6054897739504844,
|
| 1588 |
+
"grad_norm": 2.1310276875997936,
|
| 1589 |
+
"learning_rate": 4.038373542940905e-06,
|
| 1590 |
+
"loss": 0.4183,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6081808396124866,
|
| 1595 |
+
"grad_norm": 2.3017752163117744,
|
| 1596 |
+
"learning_rate": 3.992320022115492e-06,
|
| 1597 |
+
"loss": 0.4064,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6108719052744886,
|
| 1602 |
+
"grad_norm": 2.288740574605373,
|
| 1603 |
+
"learning_rate": 3.946355439143455e-06,
|
| 1604 |
+
"loss": 0.4133,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6135629709364908,
|
| 1609 |
+
"grad_norm": 2.2674805407755194,
|
| 1610 |
+
"learning_rate": 3.900483850859735e-06,
|
| 1611 |
+
"loss": 0.4275,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.616254036598493,
|
| 1616 |
+
"grad_norm": 2.3917723784081706,
|
| 1617 |
+
"learning_rate": 3.854709305891557e-06,
|
| 1618 |
+
"loss": 0.4311,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6189451022604952,
|
| 1623 |
+
"grad_norm": 2.24620365098204,
|
| 1624 |
+
"learning_rate": 3.8090358443010993e-06,
|
| 1625 |
+
"loss": 0.4111,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6216361679224973,
|
| 1630 |
+
"grad_norm": 2.297246451401738,
|
| 1631 |
+
"learning_rate": 3.7634674972289227e-06,
|
| 1632 |
+
"loss": 0.422,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6243272335844995,
|
| 1637 |
+
"grad_norm": 2.1634920709192462,
|
| 1638 |
+
"learning_rate": 3.718008286538179e-06,
|
| 1639 |
+
"loss": 0.4322,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6270182992465017,
|
| 1644 |
+
"grad_norm": 2.2583945433268857,
|
| 1645 |
+
"learning_rate": 3.67266222445964e-06,
|
| 1646 |
+
"loss": 0.4235,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6297093649085038,
|
| 1651 |
+
"grad_norm": 2.194753709445314,
|
| 1652 |
+
"learning_rate": 3.627433313237576e-06,
|
| 1653 |
+
"loss": 0.4219,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6324004305705059,
|
| 1658 |
+
"grad_norm": 2.1815873696945323,
|
| 1659 |
+
"learning_rate": 3.5823255447765233e-06,
|
| 1660 |
+
"loss": 0.4185,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.635091496232508,
|
| 1665 |
+
"grad_norm": 2.115828204783862,
|
| 1666 |
+
"learning_rate": 3.5373429002889583e-06,
|
| 1667 |
+
"loss": 0.4015,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6377825618945102,
|
| 1672 |
+
"grad_norm": 2.250398497407886,
|
| 1673 |
+
"learning_rate": 3.4924893499439096e-06,
|
| 1674 |
+
"loss": 0.4164,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6404736275565124,
|
| 1679 |
+
"grad_norm": 2.3734751418562405,
|
| 1680 |
+
"learning_rate": 3.447768852516554e-06,
|
| 1681 |
+
"loss": 0.4031,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6431646932185145,
|
| 1686 |
+
"grad_norm": 2.2971665886143056,
|
| 1687 |
+
"learning_rate": 3.4031853550388176e-06,
|
| 1688 |
+
"loss": 0.4204,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6458557588805167,
|
| 1693 |
+
"grad_norm": 2.148355296872698,
|
| 1694 |
+
"learning_rate": 3.3587427924510086e-06,
|
| 1695 |
+
"loss": 0.4088,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6485468245425189,
|
| 1700 |
+
"grad_norm": 2.2063900099924636,
|
| 1701 |
+
"learning_rate": 3.314445087254518e-06,
|
| 1702 |
+
"loss": 0.4038,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6512378902045209,
|
| 1707 |
+
"grad_norm": 2.396941304529023,
|
| 1708 |
+
"learning_rate": 3.2702961491656197e-06,
|
| 1709 |
+
"loss": 0.4095,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6539289558665231,
|
| 1714 |
+
"grad_norm": 2.4764518437414558,
|
| 1715 |
+
"learning_rate": 3.226299874770402e-06,
|
| 1716 |
+
"loss": 0.4006,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6566200215285253,
|
| 1721 |
+
"grad_norm": 2.169456697824407,
|
| 1722 |
+
"learning_rate": 3.1824601471808504e-06,
|
| 1723 |
+
"loss": 0.4102,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6593110871905274,
|
| 1728 |
+
"grad_norm": 2.2877199164126223,
|
| 1729 |
+
"learning_rate": 3.138780835692132e-06,
|
| 1730 |
+
"loss": 0.4044,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6620021528525296,
|
| 1735 |
+
"grad_norm": 2.411168783471376,
|
| 1736 |
+
"learning_rate": 3.0952657954410792e-06,
|
| 1737 |
+
"loss": 0.3998,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6646932185145318,
|
| 1742 |
+
"grad_norm": 2.1915427873990687,
|
| 1743 |
+
"learning_rate": 3.051918867065944e-06,
|
| 1744 |
+
"loss": 0.3935,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.667384284176534,
|
| 1749 |
+
"grad_norm": 2.236321241200045,
|
| 1750 |
+
"learning_rate": 3.0087438763674226e-06,
|
| 1751 |
+
"loss": 0.3968,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.670075349838536,
|
| 1756 |
+
"grad_norm": 2.153491088223765,
|
| 1757 |
+
"learning_rate": 2.9657446339709906e-06,
|
| 1758 |
+
"loss": 0.4108,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6727664155005382,
|
| 1763 |
+
"grad_norm": 2.1667604177883066,
|
| 1764 |
+
"learning_rate": 2.9229249349905686e-06,
|
| 1765 |
+
"loss": 0.4067,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6754574811625403,
|
| 1770 |
+
"grad_norm": 2.3480884088972216,
|
| 1771 |
+
"learning_rate": 2.8802885586935794e-06,
|
| 1772 |
+
"loss": 0.3973,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6781485468245425,
|
| 1777 |
+
"grad_norm": 2.363888356448008,
|
| 1778 |
+
"learning_rate": 2.837839268167373e-06,
|
| 1779 |
+
"loss": 0.3997,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6808396124865447,
|
| 1784 |
+
"grad_norm": 2.1759737776060906,
|
| 1785 |
+
"learning_rate": 2.7955808099871196e-06,
|
| 1786 |
+
"loss": 0.41,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6835306781485468,
|
| 1791 |
+
"grad_norm": 2.145769279313241,
|
| 1792 |
+
"learning_rate": 2.7535169138851124e-06,
|
| 1793 |
+
"loss": 0.3968,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.686221743810549,
|
| 1798 |
+
"grad_norm": 2.3938641882258738,
|
| 1799 |
+
"learning_rate": 2.711651292421593e-06,
|
| 1800 |
+
"loss": 0.3943,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6889128094725512,
|
| 1805 |
+
"grad_norm": 2.2985974347283435,
|
| 1806 |
+
"learning_rate": 2.6699876406570823e-06,
|
| 1807 |
+
"loss": 0.4057,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6916038751345532,
|
| 1812 |
+
"grad_norm": 2.212164701308494,
|
| 1813 |
+
"learning_rate": 2.62852963582625e-06,
|
| 1814 |
+
"loss": 0.4069,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6942949407965554,
|
| 1819 |
+
"grad_norm": 2.2332995585806783,
|
| 1820 |
+
"learning_rate": 2.5872809370133704e-06,
|
| 1821 |
+
"loss": 0.3929,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6969860064585576,
|
| 1826 |
+
"grad_norm": 2.154755198337086,
|
| 1827 |
+
"learning_rate": 2.5462451848293535e-06,
|
| 1828 |
+
"loss": 0.395,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6996770721205597,
|
| 1833 |
+
"grad_norm": 2.35147288064046,
|
| 1834 |
+
"learning_rate": 2.5054260010904423e-06,
|
| 1835 |
+
"loss": 0.4131,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7023681377825619,
|
| 1840 |
+
"grad_norm": 2.307361805622348,
|
| 1841 |
+
"learning_rate": 2.464826988498544e-06,
|
| 1842 |
+
"loss": 0.3889,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7050592034445641,
|
| 1847 |
+
"grad_norm": 2.2374159276691294,
|
| 1848 |
+
"learning_rate": 2.424451730323261e-06,
|
| 1849 |
+
"loss": 0.3911,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7077502691065662,
|
| 1854 |
+
"grad_norm": 2.0218101351051714,
|
| 1855 |
+
"learning_rate": 2.3843037900856174e-06,
|
| 1856 |
+
"loss": 0.3744,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7104413347685683,
|
| 1861 |
+
"grad_norm": 2.117430212547741,
|
| 1862 |
+
"learning_rate": 2.3443867112435585e-06,
|
| 1863 |
+
"loss": 0.3735,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7131324004305705,
|
| 1868 |
+
"grad_norm": 2.4379881536686625,
|
| 1869 |
+
"learning_rate": 2.304704016879195e-06,
|
| 1870 |
+
"loss": 0.3808,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7158234660925726,
|
| 1875 |
+
"grad_norm": 2.186063282716259,
|
| 1876 |
+
"learning_rate": 2.265259209387867e-06,
|
| 1877 |
+
"loss": 0.401,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7185145317545748,
|
| 1882 |
+
"grad_norm": 2.1034516865579316,
|
| 1883 |
+
"learning_rate": 2.226055770169002e-06,
|
| 1884 |
+
"loss": 0.3867,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.721205597416577,
|
| 1889 |
+
"grad_norm": 2.257291682766681,
|
| 1890 |
+
"learning_rate": 2.1870971593188704e-06,
|
| 1891 |
+
"loss": 0.3827,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7238966630785791,
|
| 1896 |
+
"grad_norm": 2.2641761303433983,
|
| 1897 |
+
"learning_rate": 2.148386815325179e-06,
|
| 1898 |
+
"loss": 0.3795,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7265877287405813,
|
| 1903 |
+
"grad_norm": 2.485493712722777,
|
| 1904 |
+
"learning_rate": 2.109928154763606e-06,
|
| 1905 |
+
"loss": 0.3878,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7292787944025835,
|
| 1910 |
+
"grad_norm": 2.1251911128194285,
|
| 1911 |
+
"learning_rate": 2.0717245719962347e-06,
|
| 1912 |
+
"loss": 0.3664,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7319698600645855,
|
| 1917 |
+
"grad_norm": 2.276033486452104,
|
| 1918 |
+
"learning_rate": 2.0337794388719845e-06,
|
| 1919 |
+
"loss": 0.3813,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7346609257265877,
|
| 1924 |
+
"grad_norm": 2.2995322739078747,
|
| 1925 |
+
"learning_rate": 1.9960961044290015e-06,
|
| 1926 |
+
"loss": 0.3744,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7373519913885899,
|
| 1931 |
+
"grad_norm": 2.0899009167827343,
|
| 1932 |
+
"learning_rate": 1.9586778945990785e-06,
|
| 1933 |
+
"loss": 0.3721,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.740043057050592,
|
| 1938 |
+
"grad_norm": 2.2617171211103395,
|
| 1939 |
+
"learning_rate": 1.921528111914102e-06,
|
| 1940 |
+
"loss": 0.3785,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7427341227125942,
|
| 1945 |
+
"grad_norm": 2.154826412069349,
|
| 1946 |
+
"learning_rate": 1.8846500352145753e-06,
|
| 1947 |
+
"loss": 0.3762,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7454251883745964,
|
| 1952 |
+
"grad_norm": 2.099913553338307,
|
| 1953 |
+
"learning_rate": 1.848046919360225e-06,
|
| 1954 |
+
"loss": 0.3707,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7481162540365985,
|
| 1959 |
+
"grad_norm": 2.0516253842538767,
|
| 1960 |
+
"learning_rate": 1.811721994942731e-06,
|
| 1961 |
+
"loss": 0.3744,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7508073196986006,
|
| 1966 |
+
"grad_norm": 2.1300015889778425,
|
| 1967 |
+
"learning_rate": 1.775678468000589e-06,
|
| 1968 |
+
"loss": 0.3762,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7534983853606028,
|
| 1973 |
+
"grad_norm": 2.09371262898784,
|
| 1974 |
+
"learning_rate": 1.7399195197361507e-06,
|
| 1975 |
+
"loss": 0.3767,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7561894510226049,
|
| 1980 |
+
"grad_norm": 2.0414347937365194,
|
| 1981 |
+
"learning_rate": 1.7044483062348465e-06,
|
| 1982 |
+
"loss": 0.3769,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7588805166846071,
|
| 1987 |
+
"grad_norm": 2.245862892723086,
|
| 1988 |
+
"learning_rate": 1.6692679581866334e-06,
|
| 1989 |
+
"loss": 0.3699,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7615715823466093,
|
| 1994 |
+
"grad_norm": 2.0019206442455535,
|
| 1995 |
+
"learning_rate": 1.6343815806096764e-06,
|
| 1996 |
+
"loss": 0.3718,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7642626480086114,
|
| 2001 |
+
"grad_norm": 2.0667622426417975,
|
| 2002 |
+
"learning_rate": 1.5997922525763015e-06,
|
| 2003 |
+
"loss": 0.3709,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7669537136706136,
|
| 2008 |
+
"grad_norm": 2.2454001706534252,
|
| 2009 |
+
"learning_rate": 1.5655030269412375e-06,
|
| 2010 |
+
"loss": 0.378,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7696447793326158,
|
| 2015 |
+
"grad_norm": 2.0453161314199035,
|
| 2016 |
+
"learning_rate": 1.5315169300721694e-06,
|
| 2017 |
+
"loss": 0.3798,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7723358449946178,
|
| 2022 |
+
"grad_norm": 2.122627263333417,
|
| 2023 |
+
"learning_rate": 1.4978369615826316e-06,
|
| 2024 |
+
"loss": 0.3642,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.77502691065662,
|
| 2029 |
+
"grad_norm": 2.253665103900189,
|
| 2030 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 2031 |
+
"loss": 0.3649,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7777179763186222,
|
| 2036 |
+
"grad_norm": 2.223866039627806,
|
| 2037 |
+
"learning_rate": 1.431407272839443e-06,
|
| 2038 |
+
"loss": 0.3648,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7804090419806243,
|
| 2043 |
+
"grad_norm": 2.124642169885474,
|
| 2044 |
+
"learning_rate": 1.3986634156713418e-06,
|
| 2045 |
+
"loss": 0.3686,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7831001076426265,
|
| 2050 |
+
"grad_norm": 2.039777090168692,
|
| 2051 |
+
"learning_rate": 1.3662374125363954e-06,
|
| 2052 |
+
"loss": 0.365,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7857911733046287,
|
| 2057 |
+
"grad_norm": 2.1116486617070405,
|
| 2058 |
+
"learning_rate": 1.334132125354236e-06,
|
| 2059 |
+
"loss": 0.3636,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7884822389666308,
|
| 2064 |
+
"grad_norm": 2.1955820453419217,
|
| 2065 |
+
"learning_rate": 1.302350387738101e-06,
|
| 2066 |
+
"loss": 0.3623,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7911733046286329,
|
| 2071 |
+
"grad_norm": 2.1223258884983074,
|
| 2072 |
+
"learning_rate": 1.270895004744737e-06,
|
| 2073 |
+
"loss": 0.3631,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7938643702906351,
|
| 2078 |
+
"grad_norm": 2.272320371876702,
|
| 2079 |
+
"learning_rate": 1.2397687526268248e-06,
|
| 2080 |
+
"loss": 0.3714,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7965554359526372,
|
| 2085 |
+
"grad_norm": 2.234677152395205,
|
| 2086 |
+
"learning_rate": 1.2089743785879493e-06,
|
| 2087 |
+
"loss": 0.3613,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7992465016146394,
|
| 2092 |
+
"grad_norm": 2.112632819570097,
|
| 2093 |
+
"learning_rate": 1.1785146005401292e-06,
|
| 2094 |
+
"loss": 0.3676,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8019375672766416,
|
| 2099 |
+
"grad_norm": 2.0426642476776857,
|
| 2100 |
+
"learning_rate": 1.1483921068639353e-06,
|
| 2101 |
+
"loss": 0.3589,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8046286329386437,
|
| 2106 |
+
"grad_norm": 2.2682633423265006,
|
| 2107 |
+
"learning_rate": 1.118609556171213e-06,
|
| 2108 |
+
"loss": 0.3702,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8073196986006459,
|
| 2113 |
+
"grad_norm": 2.206478926731362,
|
| 2114 |
+
"learning_rate": 1.0891695770704341e-06,
|
| 2115 |
+
"loss": 0.3669,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.810010764262648,
|
| 2120 |
+
"grad_norm": 2.0838005667062425,
|
| 2121 |
+
"learning_rate": 1.0600747679346956e-06,
|
| 2122 |
+
"loss": 0.3649,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8127018299246501,
|
| 2127 |
+
"grad_norm": 2.0526302864143515,
|
| 2128 |
+
"learning_rate": 1.0313276966723867e-06,
|
| 2129 |
+
"loss": 0.3607,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8153928955866523,
|
| 2134 |
+
"grad_norm": 2.2411019803506704,
|
| 2135 |
+
"learning_rate": 1.002930900500546e-06,
|
| 2136 |
+
"loss": 0.3608,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8180839612486545,
|
| 2141 |
+
"grad_norm": 2.2183253148113877,
|
| 2142 |
+
"learning_rate": 9.74886885720925e-07,
|
| 2143 |
+
"loss": 0.3489,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8207750269106566,
|
| 2148 |
+
"grad_norm": 2.2046741515838555,
|
| 2149 |
+
"learning_rate": 9.471981274987846e-07,
|
| 2150 |
+
"loss": 0.3582,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8234660925726588,
|
| 2155 |
+
"grad_norm": 2.0776101459474936,
|
| 2156 |
+
"learning_rate": 9.198670696444339e-07,
|
| 2157 |
+
"loss": 0.3684,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.826157158234661,
|
| 2162 |
+
"grad_norm": 2.1260490300416053,
|
| 2163 |
+
"learning_rate": 8.928961243975437e-07,
|
| 2164 |
+
"loss": 0.3602,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8288482238966631,
|
| 2169 |
+
"grad_norm": 2.203007778671738,
|
| 2170 |
+
"learning_rate": 8.662876722142327e-07,
|
| 2171 |
+
"loss": 0.3676,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8315392895586652,
|
| 2176 |
+
"grad_norm": 2.0515304374779415,
|
| 2177 |
+
"learning_rate": 8.400440615569849e-07,
|
| 2178 |
+
"loss": 0.3666,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8342303552206674,
|
| 2183 |
+
"grad_norm": 2.235903012764375,
|
| 2184 |
+
"learning_rate": 8.141676086873574e-07,
|
| 2185 |
+
"loss": 0.35,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8369214208826695,
|
| 2190 |
+
"grad_norm": 1.9898897228797456,
|
| 2191 |
+
"learning_rate": 7.886605974615574e-07,
|
| 2192 |
+
"loss": 0.3619,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8396124865446717,
|
| 2197 |
+
"grad_norm": 2.0672348209333102,
|
| 2198 |
+
"learning_rate": 7.635252791288611e-07,
|
| 2199 |
+
"loss": 0.346,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8423035522066739,
|
| 2204 |
+
"grad_norm": 2.140977921399784,
|
| 2205 |
+
"learning_rate": 7.38763872132931e-07,
|
| 2206 |
+
"loss": 0.3579,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.844994617868676,
|
| 2211 |
+
"grad_norm": 2.0050132762403137,
|
| 2212 |
+
"learning_rate": 7.143785619160026e-07,
|
| 2213 |
+
"loss": 0.3561,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8476856835306782,
|
| 2218 |
+
"grad_norm": 2.0184002747173104,
|
| 2219 |
+
"learning_rate": 6.903715007260043e-07,
|
| 2220 |
+
"loss": 0.3555,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8503767491926802,
|
| 2225 |
+
"grad_norm": 1.8477534752755447,
|
| 2226 |
+
"learning_rate": 6.667448074265954e-07,
|
| 2227 |
+
"loss": 0.3517,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8530678148546824,
|
| 2232 |
+
"grad_norm": 2.0049189975542965,
|
| 2233 |
+
"learning_rate": 6.435005673101646e-07,
|
| 2234 |
+
"loss": 0.3388,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8557588805166846,
|
| 2239 |
+
"grad_norm": 2.0424760670973345,
|
| 2240 |
+
"learning_rate": 6.206408319137703e-07,
|
| 2241 |
+
"loss": 0.371,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8584499461786868,
|
| 2246 |
+
"grad_norm": 2.0546063213646577,
|
| 2247 |
+
"learning_rate": 5.981676188380802e-07,
|
| 2248 |
+
"loss": 0.3467,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8611410118406889,
|
| 2253 |
+
"grad_norm": 2.0825364915777085,
|
| 2254 |
+
"learning_rate": 5.760829115692907e-07,
|
| 2255 |
+
"loss": 0.3593,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8638320775026911,
|
| 2260 |
+
"grad_norm": 2.276064828333872,
|
| 2261 |
+
"learning_rate": 5.543886593040737e-07,
|
| 2262 |
+
"loss": 0.3527,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8665231431646933,
|
| 2267 |
+
"grad_norm": 2.0176514994976524,
|
| 2268 |
+
"learning_rate": 5.330867767775333e-07,
|
| 2269 |
+
"loss": 0.3532,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8692142088266954,
|
| 2274 |
+
"grad_norm": 2.0708726570249922,
|
| 2275 |
+
"learning_rate": 5.121791440942131e-07,
|
| 2276 |
+
"loss": 0.3498,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8719052744886975,
|
| 2281 |
+
"grad_norm": 1.9377547369040535,
|
| 2282 |
+
"learning_rate": 4.916676065621562e-07,
|
| 2283 |
+
"loss": 0.3485,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8745963401506996,
|
| 2288 |
+
"grad_norm": 1.9361489483911087,
|
| 2289 |
+
"learning_rate": 4.715539745300429e-07,
|
| 2290 |
+
"loss": 0.3433,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8772874058127018,
|
| 2295 |
+
"grad_norm": 2.362723902179469,
|
| 2296 |
+
"learning_rate": 4.5184002322740784e-07,
|
| 2297 |
+
"loss": 0.3529,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.879978471474704,
|
| 2302 |
+
"grad_norm": 2.0061855571363765,
|
| 2303 |
+
"learning_rate": 4.3252749260795533e-07,
|
| 2304 |
+
"loss": 0.3405,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8826695371367062,
|
| 2309 |
+
"grad_norm": 2.332323171973721,
|
| 2310 |
+
"learning_rate": 4.1361808719599163e-07,
|
| 2311 |
+
"loss": 0.3522,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8853606027987083,
|
| 2316 |
+
"grad_norm": 2.1057140470170155,
|
| 2317 |
+
"learning_rate": 3.951134759359854e-07,
|
| 2318 |
+
"loss": 0.346,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8880516684607105,
|
| 2323 |
+
"grad_norm": 1.927661281496956,
|
| 2324 |
+
"learning_rate": 3.7701529204526856e-07,
|
| 2325 |
+
"loss": 0.3592,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8907427341227125,
|
| 2330 |
+
"grad_norm": 2.118098677778136,
|
| 2331 |
+
"learning_rate": 3.5932513286988436e-07,
|
| 2332 |
+
"loss": 0.3533,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8934337997847147,
|
| 2337 |
+
"grad_norm": 1.971012885096697,
|
| 2338 |
+
"learning_rate": 3.420445597436056e-07,
|
| 2339 |
+
"loss": 0.3485,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8961248654467169,
|
| 2344 |
+
"grad_norm": 2.0143344650309265,
|
| 2345 |
+
"learning_rate": 3.251750978501339e-07,
|
| 2346 |
+
"loss": 0.3503,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.898815931108719,
|
| 2351 |
+
"grad_norm": 2.0292306602899193,
|
| 2352 |
+
"learning_rate": 3.087182360884872e-07,
|
| 2353 |
+
"loss": 0.3416,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9015069967707212,
|
| 2358 |
+
"grad_norm": 1.9564839527078925,
|
| 2359 |
+
"learning_rate": 2.926754269415877e-07,
|
| 2360 |
+
"loss": 0.3507,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9041980624327234,
|
| 2365 |
+
"grad_norm": 2.1080877292592373,
|
| 2366 |
+
"learning_rate": 2.77048086348064e-07,
|
| 2367 |
+
"loss": 0.3544,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9068891280947255,
|
| 2372 |
+
"grad_norm": 2.0632107292197617,
|
| 2373 |
+
"learning_rate": 2.6183759357728543e-07,
|
| 2374 |
+
"loss": 0.3502,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9095801937567277,
|
| 2379 |
+
"grad_norm": 1.937198592937774,
|
| 2380 |
+
"learning_rate": 2.470452911076227e-07,
|
| 2381 |
+
"loss": 0.3485,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9122712594187298,
|
| 2386 |
+
"grad_norm": 2.013603885615683,
|
| 2387 |
+
"learning_rate": 2.326724845079653e-07,
|
| 2388 |
+
"loss": 0.3572,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9149623250807319,
|
| 2393 |
+
"grad_norm": 2.1281726376758066,
|
| 2394 |
+
"learning_rate": 2.1872044232248646e-07,
|
| 2395 |
+
"loss": 0.3521,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9176533907427341,
|
| 2400 |
+
"grad_norm": 2.1720549170018897,
|
| 2401 |
+
"learning_rate": 2.0519039595868706e-07,
|
| 2402 |
+
"loss": 0.355,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9203444564047363,
|
| 2407 |
+
"grad_norm": 2.058083501208556,
|
| 2408 |
+
"learning_rate": 1.9208353957870684e-07,
|
| 2409 |
+
"loss": 0.3606,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9230355220667384,
|
| 2414 |
+
"grad_norm": 2.083719674982006,
|
| 2415 |
+
"learning_rate": 1.7940102999393194e-07,
|
| 2416 |
+
"loss": 0.3548,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9257265877287406,
|
| 2421 |
+
"grad_norm": 2.130647708015,
|
| 2422 |
+
"learning_rate": 1.6714398656289154e-07,
|
| 2423 |
+
"loss": 0.3497,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9284176533907428,
|
| 2428 |
+
"grad_norm": 1.9132420017221308,
|
| 2429 |
+
"learning_rate": 1.5531349109246364e-07,
|
| 2430 |
+
"loss": 0.3356,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9311087190527448,
|
| 2435 |
+
"grad_norm": 2.0946347399647687,
|
| 2436 |
+
"learning_rate": 1.439105877423963e-07,
|
| 2437 |
+
"loss": 0.3495,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.933799784714747,
|
| 2442 |
+
"grad_norm": 2.08738192237864,
|
| 2443 |
+
"learning_rate": 1.3293628293314876e-07,
|
| 2444 |
+
"loss": 0.3517,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9364908503767492,
|
| 2449 |
+
"grad_norm": 1.9669616900562368,
|
| 2450 |
+
"learning_rate": 1.223915452570651e-07,
|
| 2451 |
+
"loss": 0.3624,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9391819160387513,
|
| 2456 |
+
"grad_norm": 2.0421605949353694,
|
| 2457 |
+
"learning_rate": 1.1227730539288717e-07,
|
| 2458 |
+
"loss": 0.3452,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9418729817007535,
|
| 2463 |
+
"grad_norm": 2.0355903627837098,
|
| 2464 |
+
"learning_rate": 1.0259445602361084e-07,
|
| 2465 |
+
"loss": 0.3492,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9445640473627557,
|
| 2470 |
+
"grad_norm": 2.0896912495997255,
|
| 2471 |
+
"learning_rate": 9.334385175769955e-08,
|
| 2472 |
+
"loss": 0.3572,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9472551130247578,
|
| 2477 |
+
"grad_norm": 2.002959531267022,
|
| 2478 |
+
"learning_rate": 8.452630905365633e-08,
|
| 2479 |
+
"loss": 0.3481,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.94994617868676,
|
| 2484 |
+
"grad_norm": 2.136820278680856,
|
| 2485 |
+
"learning_rate": 7.614260614796143e-08,
|
| 2486 |
+
"loss": 0.3435,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9526372443487621,
|
| 2491 |
+
"grad_norm": 1.9052167785885679,
|
| 2492 |
+
"learning_rate": 6.819348298638839e-08,
|
| 2493 |
+
"loss": 0.3381,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9553283100107642,
|
| 2498 |
+
"grad_norm": 2.1788907550411403,
|
| 2499 |
+
"learning_rate": 6.067964115869297e-08,
|
| 2500 |
+
"loss": 0.3577,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9580193756727664,
|
| 2505 |
+
"grad_norm": 2.085202442833548,
|
| 2506 |
+
"learning_rate": 5.36017438366937e-08,
|
| 2507 |
+
"loss": 0.3443,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9607104413347686,
|
| 2512 |
+
"grad_norm": 1.9066530306055132,
|
| 2513 |
+
"learning_rate": 4.696041571573773e-08,
|
| 2514 |
+
"loss": 0.3514,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9634015069967707,
|
| 2519 |
+
"grad_norm": 2.030414548308391,
|
| 2520 |
+
"learning_rate": 4.0756242959567596e-08,
|
| 2521 |
+
"loss": 0.3458,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9660925726587729,
|
| 2526 |
+
"grad_norm": 2.3356985672893034,
|
| 2527 |
+
"learning_rate": 3.498977314858487e-08,
|
| 2528 |
+
"loss": 0.3532,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9687836383207751,
|
| 2533 |
+
"grad_norm": 2.008020036717182,
|
| 2534 |
+
"learning_rate": 2.96615152315205e-08,
|
| 2535 |
+
"loss": 0.3371,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9714747039827771,
|
| 2540 |
+
"grad_norm": 1.996053166126216,
|
| 2541 |
+
"learning_rate": 2.4771939480516817e-08,
|
| 2542 |
+
"loss": 0.345,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9741657696447793,
|
| 2547 |
+
"grad_norm": 1.9817659128174747,
|
| 2548 |
+
"learning_rate": 2.0321477449619098e-08,
|
| 2549 |
+
"loss": 0.3428,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9768568353067815,
|
| 2554 |
+
"grad_norm": 1.968843928974531,
|
| 2555 |
+
"learning_rate": 1.6310521936688806e-08,
|
| 2556 |
+
"loss": 0.3475,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9795479009687836,
|
| 2561 |
+
"grad_norm": 2.008709148033631,
|
| 2562 |
+
"learning_rate": 1.2739426948732426e-08,
|
| 2563 |
+
"loss": 0.3579,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9822389666307858,
|
| 2568 |
+
"grad_norm": 2.0860175169319852,
|
| 2569 |
+
"learning_rate": 9.608507670659239e-09,
|
| 2570 |
+
"loss": 0.3269,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.984930032292788,
|
| 2575 |
+
"grad_norm": 2.1919123246360415,
|
| 2576 |
+
"learning_rate": 6.918040437463025e-09,
|
| 2577 |
+
"loss": 0.3546,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9876210979547901,
|
| 2582 |
+
"grad_norm": 2.197666814347832,
|
| 2583 |
+
"learning_rate": 4.668262709830451e-09,
|
| 2584 |
+
"loss": 0.3476,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9903121636167922,
|
| 2589 |
+
"grad_norm": 2.0435447242057716,
|
| 2590 |
+
"learning_rate": 2.8593730531861764e-09,
|
| 2591 |
+
"loss": 0.3309,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9930032292787944,
|
| 2596 |
+
"grad_norm": 2.1036408486178244,
|
| 2597 |
+
"learning_rate": 1.4915311201635362e-09,
|
| 2598 |
+
"loss": 0.3428,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9956942949407965,
|
| 2603 |
+
"grad_norm": 1.9217810613809283,
|
| 2604 |
+
"learning_rate": 5.648576365169245e-10,
|
| 2605 |
+
"loss": 0.3392,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9983853606027987,
|
| 2610 |
+
"grad_norm": 2.1168830996512846,
|
| 2611 |
+
"learning_rate": 7.943439046531609e-11,
|
| 2612 |
+
"loss": 0.3501,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 1.0,
|
| 2617 |
+
"eval_runtime": 3.853,
|
| 2618 |
+
"eval_samples_per_second": 2.595,
|
| 2619 |
+
"eval_steps_per_second": 0.779,
|
| 2620 |
+
"step": 1858
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 1.0,
|
| 2624 |
+
"step": 1858,
|
| 2625 |
+
"total_flos": 194513700126720.0,
|
| 2626 |
+
"train_loss": 0.5110224842126967,
|
| 2627 |
+
"train_runtime": 16738.0887,
|
| 2628 |
+
"train_samples_per_second": 1.776,
|
| 2629 |
+
"train_steps_per_second": 0.111
|
| 2630 |
+
}
|
| 2631 |
+
],
|
| 2632 |
+
"logging_steps": 5,
|
| 2633 |
+
"max_steps": 1858,
|
| 2634 |
+
"num_input_tokens_seen": 0,
|
| 2635 |
+
"num_train_epochs": 1,
|
| 2636 |
+
"save_steps": 100,
|
| 2637 |
+
"stateful_callbacks": {
|
| 2638 |
+
"TrainerControl": {
|
| 2639 |
+
"args": {
|
| 2640 |
+
"should_epoch_stop": false,
|
| 2641 |
+
"should_evaluate": false,
|
| 2642 |
+
"should_log": false,
|
| 2643 |
+
"should_save": true,
|
| 2644 |
+
"should_training_stop": true
|
| 2645 |
+
},
|
| 2646 |
+
"attributes": {}
|
| 2647 |
+
}
|
| 2648 |
+
},
|
| 2649 |
+
"total_flos": 194513700126720.0,
|
| 2650 |
+
"train_batch_size": 4,
|
| 2651 |
+
"trial_name": null,
|
| 2652 |
+
"trial_params": null
|
| 2653 |
+
}
|