Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep23_06-22-41_COE-CS-sv003/events.out.tfevents.1727072868.COE-CS-sv003.732459.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2674 -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_stylemixbatch16_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_stylemixbatch16_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": 195874667888640.0,
|
| 4 |
+
"train_loss": 0.5146575008465732,
|
| 5 |
+
"train_runtime": 16798.2789,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.782,
|
| 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_06-22-41_COE-CS-sv003/events.out.tfevents.1727072868.COE-CS-sv003.732459.0
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:d33f46d19bed050ccf0e4fc78ba5d25e1bed4f87f234686543fa724b089c5667
|
| 3 |
+
size 84892
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 195874667888640.0,
|
| 4 |
+
"train_loss": 0.5146575008465732,
|
| 5 |
+
"train_runtime": 16798.2789,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.782,
|
| 8 |
+
"train_steps_per_second": 0.111
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2674 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1871,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005344735435595938,
|
| 13 |
+
"grad_norm": 23.61518103763694,
|
| 14 |
+
"learning_rate": 5.319148936170213e-08,
|
| 15 |
+
"loss": 1.3621,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002672367717797969,
|
| 20 |
+
"grad_norm": 23.52338386367116,
|
| 21 |
+
"learning_rate": 2.6595744680851066e-07,
|
| 22 |
+
"loss": 1.3454,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005344735435595938,
|
| 27 |
+
"grad_norm": 15.902737248388277,
|
| 28 |
+
"learning_rate": 5.319148936170213e-07,
|
| 29 |
+
"loss": 1.3168,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008017103153393906,
|
| 34 |
+
"grad_norm": 11.62038726971463,
|
| 35 |
+
"learning_rate": 7.97872340425532e-07,
|
| 36 |
+
"loss": 1.1663,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010689470871191877,
|
| 41 |
+
"grad_norm": 10.50913201186014,
|
| 42 |
+
"learning_rate": 1.0638297872340427e-06,
|
| 43 |
+
"loss": 1.0566,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013361838588989846,
|
| 48 |
+
"grad_norm": 3.560582254809462,
|
| 49 |
+
"learning_rate": 1.3297872340425533e-06,
|
| 50 |
+
"loss": 0.9328,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.016034206306787813,
|
| 55 |
+
"grad_norm": 3.318390385053047,
|
| 56 |
+
"learning_rate": 1.595744680851064e-06,
|
| 57 |
+
"loss": 0.8966,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.01870657402458578,
|
| 62 |
+
"grad_norm": 2.9993400845793654,
|
| 63 |
+
"learning_rate": 1.8617021276595745e-06,
|
| 64 |
+
"loss": 0.8769,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.021378941742383754,
|
| 69 |
+
"grad_norm": 2.9199745023125967,
|
| 70 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 71 |
+
"loss": 0.8607,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.024051309460181722,
|
| 76 |
+
"grad_norm": 2.9991036135574545,
|
| 77 |
+
"learning_rate": 2.393617021276596e-06,
|
| 78 |
+
"loss": 0.8516,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02672367717797969,
|
| 83 |
+
"grad_norm": 2.834663670092906,
|
| 84 |
+
"learning_rate": 2.6595744680851065e-06,
|
| 85 |
+
"loss": 0.8328,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02939604489577766,
|
| 90 |
+
"grad_norm": 3.071674166085022,
|
| 91 |
+
"learning_rate": 2.9255319148936174e-06,
|
| 92 |
+
"loss": 0.8212,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.032068412613575625,
|
| 97 |
+
"grad_norm": 3.0481603346640704,
|
| 98 |
+
"learning_rate": 3.191489361702128e-06,
|
| 99 |
+
"loss": 0.8242,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.034740780331373594,
|
| 104 |
+
"grad_norm": 3.0857566743029907,
|
| 105 |
+
"learning_rate": 3.457446808510639e-06,
|
| 106 |
+
"loss": 0.8045,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03741314804917156,
|
| 111 |
+
"grad_norm": 3.2871683892802004,
|
| 112 |
+
"learning_rate": 3.723404255319149e-06,
|
| 113 |
+
"loss": 0.7966,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04008551576696953,
|
| 118 |
+
"grad_norm": 3.2034895167525277,
|
| 119 |
+
"learning_rate": 3.98936170212766e-06,
|
| 120 |
+
"loss": 0.7806,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04275788348476751,
|
| 125 |
+
"grad_norm": 3.226417198218916,
|
| 126 |
+
"learning_rate": 4.255319148936171e-06,
|
| 127 |
+
"loss": 0.7729,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.045430251202565476,
|
| 132 |
+
"grad_norm": 3.170786690584355,
|
| 133 |
+
"learning_rate": 4.521276595744681e-06,
|
| 134 |
+
"loss": 0.7828,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.048102618920363445,
|
| 139 |
+
"grad_norm": 3.0895839104025717,
|
| 140 |
+
"learning_rate": 4.787234042553192e-06,
|
| 141 |
+
"loss": 0.7667,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.050774986638161414,
|
| 146 |
+
"grad_norm": 3.192938409188057,
|
| 147 |
+
"learning_rate": 5.053191489361703e-06,
|
| 148 |
+
"loss": 0.7595,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05344735435595938,
|
| 153 |
+
"grad_norm": 3.206609902202038,
|
| 154 |
+
"learning_rate": 5.319148936170213e-06,
|
| 155 |
+
"loss": 0.7555,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05611972207375735,
|
| 160 |
+
"grad_norm": 3.175589519512037,
|
| 161 |
+
"learning_rate": 5.5851063829787235e-06,
|
| 162 |
+
"loss": 0.7457,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05879208979155532,
|
| 167 |
+
"grad_norm": 3.0405783301327975,
|
| 168 |
+
"learning_rate": 5.851063829787235e-06,
|
| 169 |
+
"loss": 0.7452,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06146445750935329,
|
| 174 |
+
"grad_norm": 2.8250036844985584,
|
| 175 |
+
"learning_rate": 6.117021276595745e-06,
|
| 176 |
+
"loss": 0.7378,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06413682522715125,
|
| 181 |
+
"grad_norm": 3.1222586810818775,
|
| 182 |
+
"learning_rate": 6.382978723404256e-06,
|
| 183 |
+
"loss": 0.7337,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06680919294494922,
|
| 188 |
+
"grad_norm": 2.8888701910636425,
|
| 189 |
+
"learning_rate": 6.648936170212767e-06,
|
| 190 |
+
"loss": 0.7321,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06948156066274719,
|
| 195 |
+
"grad_norm": 3.0481545986847034,
|
| 196 |
+
"learning_rate": 6.914893617021278e-06,
|
| 197 |
+
"loss": 0.7356,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07215392838054516,
|
| 202 |
+
"grad_norm": 2.9221465817003316,
|
| 203 |
+
"learning_rate": 7.1808510638297875e-06,
|
| 204 |
+
"loss": 0.7163,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07482629609834313,
|
| 209 |
+
"grad_norm": 2.823797619258334,
|
| 210 |
+
"learning_rate": 7.446808510638298e-06,
|
| 211 |
+
"loss": 0.7179,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0774986638161411,
|
| 216 |
+
"grad_norm": 2.974797988982762,
|
| 217 |
+
"learning_rate": 7.71276595744681e-06,
|
| 218 |
+
"loss": 0.7083,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08017103153393906,
|
| 223 |
+
"grad_norm": 3.1868031201090252,
|
| 224 |
+
"learning_rate": 7.97872340425532e-06,
|
| 225 |
+
"loss": 0.7223,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08284339925173703,
|
| 230 |
+
"grad_norm": 3.313855919178953,
|
| 231 |
+
"learning_rate": 8.24468085106383e-06,
|
| 232 |
+
"loss": 0.7223,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08551576696953501,
|
| 237 |
+
"grad_norm": 2.917840787459801,
|
| 238 |
+
"learning_rate": 8.510638297872341e-06,
|
| 239 |
+
"loss": 0.7195,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08818813468733298,
|
| 244 |
+
"grad_norm": 2.770307352801897,
|
| 245 |
+
"learning_rate": 8.776595744680852e-06,
|
| 246 |
+
"loss": 0.7235,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09086050240513095,
|
| 251 |
+
"grad_norm": 2.765197627364069,
|
| 252 |
+
"learning_rate": 9.042553191489362e-06,
|
| 253 |
+
"loss": 0.7094,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09353287012292892,
|
| 258 |
+
"grad_norm": 2.9219350077214283,
|
| 259 |
+
"learning_rate": 9.308510638297872e-06,
|
| 260 |
+
"loss": 0.6998,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09620523784072689,
|
| 265 |
+
"grad_norm": 3.1415213219167293,
|
| 266 |
+
"learning_rate": 9.574468085106385e-06,
|
| 267 |
+
"loss": 0.7102,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09887760555852486,
|
| 272 |
+
"grad_norm": 2.7242573109350645,
|
| 273 |
+
"learning_rate": 9.840425531914895e-06,
|
| 274 |
+
"loss": 0.7228,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10154997327632283,
|
| 279 |
+
"grad_norm": 2.859985126301972,
|
| 280 |
+
"learning_rate": 9.99996515575899e-06,
|
| 281 |
+
"loss": 0.7254,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1042223409941208,
|
| 286 |
+
"grad_norm": 2.665798747814334,
|
| 287 |
+
"learning_rate": 9.99957316362496e-06,
|
| 288 |
+
"loss": 0.7109,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10689470871191876,
|
| 293 |
+
"grad_norm": 2.9076197629295817,
|
| 294 |
+
"learning_rate": 9.998745658315924e-06,
|
| 295 |
+
"loss": 0.6989,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10956707642971673,
|
| 300 |
+
"grad_norm": 2.8125731981919047,
|
| 301 |
+
"learning_rate": 9.997482711915926e-06,
|
| 302 |
+
"loss": 0.718,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1122394441475147,
|
| 307 |
+
"grad_norm": 2.6725574630640594,
|
| 308 |
+
"learning_rate": 9.99578443444032e-06,
|
| 309 |
+
"loss": 0.7112,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11491181186531267,
|
| 314 |
+
"grad_norm": 2.8593331201166152,
|
| 315 |
+
"learning_rate": 9.993650973826177e-06,
|
| 316 |
+
"loss": 0.7091,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11758417958311064,
|
| 321 |
+
"grad_norm": 2.6883306424483506,
|
| 322 |
+
"learning_rate": 9.991082515919402e-06,
|
| 323 |
+
"loss": 0.6982,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.12025654730090861,
|
| 328 |
+
"grad_norm": 2.786493474222006,
|
| 329 |
+
"learning_rate": 9.988079284458547e-06,
|
| 330 |
+
"loss": 0.6956,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12292891501870658,
|
| 335 |
+
"grad_norm": 2.576508494305329,
|
| 336 |
+
"learning_rate": 9.98464154105532e-06,
|
| 337 |
+
"loss": 0.6903,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12560128273650453,
|
| 342 |
+
"grad_norm": 2.7285477172023342,
|
| 343 |
+
"learning_rate": 9.980769585171795e-06,
|
| 344 |
+
"loss": 0.69,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1282736504543025,
|
| 349 |
+
"grad_norm": 2.6675103293265545,
|
| 350 |
+
"learning_rate": 9.976463754094321e-06,
|
| 351 |
+
"loss": 0.6978,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.13094601817210047,
|
| 356 |
+
"grad_norm": 2.7330300721548664,
|
| 357 |
+
"learning_rate": 9.971724422904154e-06,
|
| 358 |
+
"loss": 0.6841,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13361838588989844,
|
| 363 |
+
"grad_norm": 2.7271057104317804,
|
| 364 |
+
"learning_rate": 9.966552004444772e-06,
|
| 365 |
+
"loss": 0.6969,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1362907536076964,
|
| 370 |
+
"grad_norm": 2.578197568780461,
|
| 371 |
+
"learning_rate": 9.960946949285915e-06,
|
| 372 |
+
"loss": 0.6991,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13896312132549438,
|
| 377 |
+
"grad_norm": 2.610755637023348,
|
| 378 |
+
"learning_rate": 9.954909745684339e-06,
|
| 379 |
+
"loss": 0.7001,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14163548904329235,
|
| 384 |
+
"grad_norm": 2.656700773061203,
|
| 385 |
+
"learning_rate": 9.948440919541277e-06,
|
| 386 |
+
"loss": 0.6793,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14430785676109031,
|
| 391 |
+
"grad_norm": 2.714010328000571,
|
| 392 |
+
"learning_rate": 9.94154103435664e-06,
|
| 393 |
+
"loss": 0.6854,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14698022447888828,
|
| 398 |
+
"grad_norm": 2.5795427101760686,
|
| 399 |
+
"learning_rate": 9.934210691179918e-06,
|
| 400 |
+
"loss": 0.6705,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14965259219668625,
|
| 405 |
+
"grad_norm": 2.644406518293481,
|
| 406 |
+
"learning_rate": 9.926450528557828e-06,
|
| 407 |
+
"loss": 0.6872,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.15232495991448422,
|
| 412 |
+
"grad_norm": 2.626026852267855,
|
| 413 |
+
"learning_rate": 9.918261222478687e-06,
|
| 414 |
+
"loss": 0.6921,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1549973276322822,
|
| 419 |
+
"grad_norm": 2.675148415559724,
|
| 420 |
+
"learning_rate": 9.909643486313533e-06,
|
| 421 |
+
"loss": 0.6629,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15766969535008016,
|
| 426 |
+
"grad_norm": 2.6288509142094734,
|
| 427 |
+
"learning_rate": 9.900598070753981e-06,
|
| 428 |
+
"loss": 0.6742,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.16034206306787813,
|
| 433 |
+
"grad_norm": 2.607213205803097,
|
| 434 |
+
"learning_rate": 9.891125763746824e-06,
|
| 435 |
+
"loss": 0.6734,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.1630144307856761,
|
| 440 |
+
"grad_norm": 2.699217392639012,
|
| 441 |
+
"learning_rate": 9.881227390425404e-06,
|
| 442 |
+
"loss": 0.6725,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16568679850347406,
|
| 447 |
+
"grad_norm": 2.500608790665497,
|
| 448 |
+
"learning_rate": 9.87090381303772e-06,
|
| 449 |
+
"loss": 0.6782,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16835916622127206,
|
| 454 |
+
"grad_norm": 2.4988371088586443,
|
| 455 |
+
"learning_rate": 9.860155930871341e-06,
|
| 456 |
+
"loss": 0.6643,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.17103153393907003,
|
| 461 |
+
"grad_norm": 2.49890305345079,
|
| 462 |
+
"learning_rate": 9.848984680175049e-06,
|
| 463 |
+
"loss": 0.6561,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.173703901656868,
|
| 468 |
+
"grad_norm": 2.4436899486571733,
|
| 469 |
+
"learning_rate": 9.837391034077286e-06,
|
| 470 |
+
"loss": 0.6623,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17637626937466597,
|
| 475 |
+
"grad_norm": 2.4799571818750605,
|
| 476 |
+
"learning_rate": 9.825376002501393e-06,
|
| 477 |
+
"loss": 0.6649,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17904863709246394,
|
| 482 |
+
"grad_norm": 2.5658352982346204,
|
| 483 |
+
"learning_rate": 9.812940632077629e-06,
|
| 484 |
+
"loss": 0.6666,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.1817210048102619,
|
| 489 |
+
"grad_norm": 2.5038692575935007,
|
| 490 |
+
"learning_rate": 9.800086006051996e-06,
|
| 491 |
+
"loss": 0.6681,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18439337252805987,
|
| 496 |
+
"grad_norm": 2.4140997941308586,
|
| 497 |
+
"learning_rate": 9.786813244191885e-06,
|
| 498 |
+
"loss": 0.6603,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18706574024585784,
|
| 503 |
+
"grad_norm": 2.4461106422469387,
|
| 504 |
+
"learning_rate": 9.773123502688532e-06,
|
| 505 |
+
"loss": 0.6615,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.1897381079636558,
|
| 510 |
+
"grad_norm": 2.4373544953494926,
|
| 511 |
+
"learning_rate": 9.759017974056292e-06,
|
| 512 |
+
"loss": 0.6496,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.19241047568145378,
|
| 517 |
+
"grad_norm": 2.48009393879711,
|
| 518 |
+
"learning_rate": 9.744497887028774e-06,
|
| 519 |
+
"loss": 0.6507,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19508284339925175,
|
| 524 |
+
"grad_norm": 2.420896783498181,
|
| 525 |
+
"learning_rate": 9.729564506451791e-06,
|
| 526 |
+
"loss": 0.6499,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19775521111704972,
|
| 531 |
+
"grad_norm": 2.4342500952426223,
|
| 532 |
+
"learning_rate": 9.714219133173194e-06,
|
| 533 |
+
"loss": 0.6545,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.20042757883484769,
|
| 538 |
+
"grad_norm": 2.726231096446901,
|
| 539 |
+
"learning_rate": 9.698463103929542e-06,
|
| 540 |
+
"loss": 0.6384,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.20309994655264565,
|
| 545 |
+
"grad_norm": 2.365009266839175,
|
| 546 |
+
"learning_rate": 9.682297791229668e-06,
|
| 547 |
+
"loss": 0.6583,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20577231427044362,
|
| 552 |
+
"grad_norm": 2.390159536711186,
|
| 553 |
+
"learning_rate": 9.665724603235115e-06,
|
| 554 |
+
"loss": 0.6303,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2084446819882416,
|
| 559 |
+
"grad_norm": 2.527701453079277,
|
| 560 |
+
"learning_rate": 9.648744983637471e-06,
|
| 561 |
+
"loss": 0.6569,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.21111704970603956,
|
| 566 |
+
"grad_norm": 2.417799402235962,
|
| 567 |
+
"learning_rate": 9.631360411532609e-06,
|
| 568 |
+
"loss": 0.6325,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.21378941742383753,
|
| 573 |
+
"grad_norm": 2.4529437190653884,
|
| 574 |
+
"learning_rate": 9.61357240129185e-06,
|
| 575 |
+
"loss": 0.6577,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2164617851416355,
|
| 580 |
+
"grad_norm": 2.2689373207444725,
|
| 581 |
+
"learning_rate": 9.59538250243003e-06,
|
| 582 |
+
"loss": 0.6297,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.21913415285943347,
|
| 587 |
+
"grad_norm": 2.5536672740881223,
|
| 588 |
+
"learning_rate": 9.576792299470537e-06,
|
| 589 |
+
"loss": 0.6344,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.22180652057723144,
|
| 594 |
+
"grad_norm": 2.592880004163201,
|
| 595 |
+
"learning_rate": 9.557803411807283e-06,
|
| 596 |
+
"loss": 0.6298,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2244788882950294,
|
| 601 |
+
"grad_norm": 2.3977529842883167,
|
| 602 |
+
"learning_rate": 9.538417493563621e-06,
|
| 603 |
+
"loss": 0.6235,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22715125601282737,
|
| 608 |
+
"grad_norm": 2.409144104508164,
|
| 609 |
+
"learning_rate": 9.518636233448276e-06,
|
| 610 |
+
"loss": 0.6312,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.22982362373062534,
|
| 615 |
+
"grad_norm": 2.345148626579648,
|
| 616 |
+
"learning_rate": 9.498461354608228e-06,
|
| 617 |
+
"loss": 0.6335,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2324959914484233,
|
| 622 |
+
"grad_norm": 2.7130782463024823,
|
| 623 |
+
"learning_rate": 9.47789461447861e-06,
|
| 624 |
+
"loss": 0.6402,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23516835916622128,
|
| 629 |
+
"grad_norm": 2.4306200930423363,
|
| 630 |
+
"learning_rate": 9.456937804629623e-06,
|
| 631 |
+
"loss": 0.6199,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23784072688401925,
|
| 636 |
+
"grad_norm": 2.3667962521395154,
|
| 637 |
+
"learning_rate": 9.435592750610469e-06,
|
| 638 |
+
"loss": 0.6249,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.24051309460181722,
|
| 643 |
+
"grad_norm": 2.3445164292760503,
|
| 644 |
+
"learning_rate": 9.413861311790327e-06,
|
| 645 |
+
"loss": 0.635,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2431854623196152,
|
| 650 |
+
"grad_norm": 2.408243915012667,
|
| 651 |
+
"learning_rate": 9.391745381196382e-06,
|
| 652 |
+
"loss": 0.6199,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24585783003741316,
|
| 657 |
+
"grad_norm": 2.7759339062456228,
|
| 658 |
+
"learning_rate": 9.369246885348926e-06,
|
| 659 |
+
"loss": 0.6226,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.24853019775521112,
|
| 664 |
+
"grad_norm": 2.3972288359532863,
|
| 665 |
+
"learning_rate": 9.346367784093538e-06,
|
| 666 |
+
"loss": 0.616,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.25120256547300907,
|
| 671 |
+
"grad_norm": 2.4374485416456135,
|
| 672 |
+
"learning_rate": 9.32311007043036e-06,
|
| 673 |
+
"loss": 0.6081,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.25387493319080706,
|
| 678 |
+
"grad_norm": 2.3942250427385177,
|
| 679 |
+
"learning_rate": 9.299475770340492e-06,
|
| 680 |
+
"loss": 0.6166,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.256547300908605,
|
| 685 |
+
"grad_norm": 2.4289153537652948,
|
| 686 |
+
"learning_rate": 9.275466942609495e-06,
|
| 687 |
+
"loss": 0.6334,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.259219668626403,
|
| 692 |
+
"grad_norm": 2.367943437037186,
|
| 693 |
+
"learning_rate": 9.251085678648072e-06,
|
| 694 |
+
"loss": 0.608,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.26189203634420094,
|
| 699 |
+
"grad_norm": 2.498532655161112,
|
| 700 |
+
"learning_rate": 9.226334102309862e-06,
|
| 701 |
+
"loss": 0.6058,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.26456440406199894,
|
| 706 |
+
"grad_norm": 2.390835195956499,
|
| 707 |
+
"learning_rate": 9.201214369706448e-06,
|
| 708 |
+
"loss": 0.6195,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2672367717797969,
|
| 713 |
+
"grad_norm": 2.3875744730291304,
|
| 714 |
+
"learning_rate": 9.17572866901953e-06,
|
| 715 |
+
"loss": 0.6111,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2699091394975949,
|
| 720 |
+
"grad_norm": 2.433008431887238,
|
| 721 |
+
"learning_rate": 9.14987922031031e-06,
|
| 722 |
+
"loss": 0.6147,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2725815072153928,
|
| 727 |
+
"grad_norm": 2.304360833496131,
|
| 728 |
+
"learning_rate": 9.123668275326113e-06,
|
| 729 |
+
"loss": 0.6007,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2752538749331908,
|
| 734 |
+
"grad_norm": 2.455168124864067,
|
| 735 |
+
"learning_rate": 9.097098117304223e-06,
|
| 736 |
+
"loss": 0.5946,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.27792624265098875,
|
| 741 |
+
"grad_norm": 2.3757853193488483,
|
| 742 |
+
"learning_rate": 9.070171060773007e-06,
|
| 743 |
+
"loss": 0.5796,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.28059861036878675,
|
| 748 |
+
"grad_norm": 2.438156658306187,
|
| 749 |
+
"learning_rate": 9.042889451350274e-06,
|
| 750 |
+
"loss": 0.6095,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2832709780865847,
|
| 755 |
+
"grad_norm": 2.3933478156476267,
|
| 756 |
+
"learning_rate": 9.015255665538972e-06,
|
| 757 |
+
"loss": 0.6138,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2859433458043827,
|
| 762 |
+
"grad_norm": 2.4102250220370918,
|
| 763 |
+
"learning_rate": 8.987272110520154e-06,
|
| 764 |
+
"loss": 0.5949,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.28861571352218063,
|
| 769 |
+
"grad_norm": 2.3405736149434766,
|
| 770 |
+
"learning_rate": 8.958941223943292e-06,
|
| 771 |
+
"loss": 0.609,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2912880812399786,
|
| 776 |
+
"grad_norm": 2.508552589080363,
|
| 777 |
+
"learning_rate": 8.930265473713939e-06,
|
| 778 |
+
"loss": 0.6083,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.29396044895777657,
|
| 783 |
+
"grad_norm": 2.362031389976222,
|
| 784 |
+
"learning_rate": 8.901247357778742e-06,
|
| 785 |
+
"loss": 0.5972,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.29663281667557456,
|
| 790 |
+
"grad_norm": 2.400704295368671,
|
| 791 |
+
"learning_rate": 8.871889403907853e-06,
|
| 792 |
+
"loss": 0.5953,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.2993051843933725,
|
| 797 |
+
"grad_norm": 2.4672308114305914,
|
| 798 |
+
"learning_rate": 8.842194169474727e-06,
|
| 799 |
+
"loss": 0.5975,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3019775521111705,
|
| 804 |
+
"grad_norm": 2.4217091412097798,
|
| 805 |
+
"learning_rate": 8.812164241233354e-06,
|
| 806 |
+
"loss": 0.5865,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.30464991982896844,
|
| 811 |
+
"grad_norm": 2.4016221326031246,
|
| 812 |
+
"learning_rate": 8.781802235092927e-06,
|
| 813 |
+
"loss": 0.5853,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.30732228754676644,
|
| 818 |
+
"grad_norm": 2.4448063152595947,
|
| 819 |
+
"learning_rate": 8.751110795889966e-06,
|
| 820 |
+
"loss": 0.5846,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3099946552645644,
|
| 825 |
+
"grad_norm": 2.4101255235456365,
|
| 826 |
+
"learning_rate": 8.72009259715793e-06,
|
| 827 |
+
"loss": 0.5826,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3126670229823624,
|
| 832 |
+
"grad_norm": 2.3229216437391367,
|
| 833 |
+
"learning_rate": 8.688750340894324e-06,
|
| 834 |
+
"loss": 0.6094,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3153393907001603,
|
| 839 |
+
"grad_norm": 2.364943931043124,
|
| 840 |
+
"learning_rate": 8.657086757325328e-06,
|
| 841 |
+
"loss": 0.5968,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3180117584179583,
|
| 846 |
+
"grad_norm": 2.4893765104704952,
|
| 847 |
+
"learning_rate": 8.625104604667965e-06,
|
| 848 |
+
"loss": 0.5721,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.32068412613575625,
|
| 853 |
+
"grad_norm": 2.4524834063054386,
|
| 854 |
+
"learning_rate": 8.592806668889835e-06,
|
| 855 |
+
"loss": 0.567,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.32335649385355425,
|
| 860 |
+
"grad_norm": 2.4028879963718324,
|
| 861 |
+
"learning_rate": 8.560195763466428e-06,
|
| 862 |
+
"loss": 0.5931,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3260288615713522,
|
| 867 |
+
"grad_norm": 2.3843093171732157,
|
| 868 |
+
"learning_rate": 8.527274729136042e-06,
|
| 869 |
+
"loss": 0.5843,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3287012292891502,
|
| 874 |
+
"grad_norm": 2.3502440673042577,
|
| 875 |
+
"learning_rate": 8.494046433652327e-06,
|
| 876 |
+
"loss": 0.5738,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.33137359700694813,
|
| 881 |
+
"grad_norm": 2.284954690172926,
|
| 882 |
+
"learning_rate": 8.460513771534475e-06,
|
| 883 |
+
"loss": 0.559,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3340459647247461,
|
| 888 |
+
"grad_norm": 2.3612010873776414,
|
| 889 |
+
"learning_rate": 8.426679663815073e-06,
|
| 890 |
+
"loss": 0.5918,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3367183324425441,
|
| 895 |
+
"grad_norm": 2.161417560267832,
|
| 896 |
+
"learning_rate": 8.392547057785662e-06,
|
| 897 |
+
"loss": 0.5483,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.33939070016034206,
|
| 902 |
+
"grad_norm": 2.4200810566750306,
|
| 903 |
+
"learning_rate": 8.358118926739984e-06,
|
| 904 |
+
"loss": 0.5514,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.34206306787814006,
|
| 909 |
+
"grad_norm": 2.497752095251993,
|
| 910 |
+
"learning_rate": 8.323398269714994e-06,
|
| 911 |
+
"loss": 0.5798,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.344735435595938,
|
| 916 |
+
"grad_norm": 2.4247160367906413,
|
| 917 |
+
"learning_rate": 8.288388111229601e-06,
|
| 918 |
+
"loss": 0.5579,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.347407803313736,
|
| 923 |
+
"grad_norm": 2.4629425257040034,
|
| 924 |
+
"learning_rate": 8.25309150102121e-06,
|
| 925 |
+
"loss": 0.5599,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.35008017103153394,
|
| 930 |
+
"grad_norm": 2.3769509731214855,
|
| 931 |
+
"learning_rate": 8.217511513780056e-06,
|
| 932 |
+
"loss": 0.5806,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.35275253874933193,
|
| 937 |
+
"grad_norm": 2.327564053302344,
|
| 938 |
+
"learning_rate": 8.181651248881364e-06,
|
| 939 |
+
"loss": 0.5533,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3554249064671299,
|
| 944 |
+
"grad_norm": 2.621051043502927,
|
| 945 |
+
"learning_rate": 8.145513830115367e-06,
|
| 946 |
+
"loss": 0.5521,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.35809727418492787,
|
| 951 |
+
"grad_norm": 2.2454195387127065,
|
| 952 |
+
"learning_rate": 8.109102405415195e-06,
|
| 953 |
+
"loss": 0.5569,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3607696419027258,
|
| 958 |
+
"grad_norm": 2.4656771958068373,
|
| 959 |
+
"learning_rate": 8.072420146582649e-06,
|
| 960 |
+
"loss": 0.5661,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3634420096205238,
|
| 965 |
+
"grad_norm": 2.3668021071793977,
|
| 966 |
+
"learning_rate": 8.035470249011916e-06,
|
| 967 |
+
"loss": 0.5713,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.36611437733832175,
|
| 972 |
+
"grad_norm": 2.461540400642502,
|
| 973 |
+
"learning_rate": 7.998255931411208e-06,
|
| 974 |
+
"loss": 0.548,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.36878674505611975,
|
| 979 |
+
"grad_norm": 2.3787889057132805,
|
| 980 |
+
"learning_rate": 7.960780435522387e-06,
|
| 981 |
+
"loss": 0.5583,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.3714591127739177,
|
| 986 |
+
"grad_norm": 2.43169471297202,
|
| 987 |
+
"learning_rate": 7.923047025838573e-06,
|
| 988 |
+
"loss": 0.5624,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3741314804917157,
|
| 993 |
+
"grad_norm": 2.5825683747421486,
|
| 994 |
+
"learning_rate": 7.885058989319776e-06,
|
| 995 |
+
"loss": 0.5597,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3768038482095136,
|
| 1000 |
+
"grad_norm": 2.2942237910435876,
|
| 1001 |
+
"learning_rate": 7.846819635106569e-06,
|
| 1002 |
+
"loss": 0.5468,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3794762159273116,
|
| 1007 |
+
"grad_norm": 2.309892265310505,
|
| 1008 |
+
"learning_rate": 7.808332294231824e-06,
|
| 1009 |
+
"loss": 0.5421,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.38214858364510956,
|
| 1014 |
+
"grad_norm": 2.5953061845305445,
|
| 1015 |
+
"learning_rate": 7.769600319330553e-06,
|
| 1016 |
+
"loss": 0.5466,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.38482095136290756,
|
| 1021 |
+
"grad_norm": 2.290660724345896,
|
| 1022 |
+
"learning_rate": 7.73062708434785e-06,
|
| 1023 |
+
"loss": 0.5549,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3874933190807055,
|
| 1028 |
+
"grad_norm": 2.4134172892827466,
|
| 1029 |
+
"learning_rate": 7.691415984244998e-06,
|
| 1030 |
+
"loss": 0.5573,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.3901656867985035,
|
| 1035 |
+
"grad_norm": 2.43710642014278,
|
| 1036 |
+
"learning_rate": 7.651970434703724e-06,
|
| 1037 |
+
"loss": 0.528,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.39283805451630144,
|
| 1042 |
+
"grad_norm": 2.4549602694531165,
|
| 1043 |
+
"learning_rate": 7.612293871828662e-06,
|
| 1044 |
+
"loss": 0.5209,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.39551042223409943,
|
| 1049 |
+
"grad_norm": 2.472644139489206,
|
| 1050 |
+
"learning_rate": 7.572389751848037e-06,
|
| 1051 |
+
"loss": 0.5274,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.3981827899518974,
|
| 1056 |
+
"grad_norm": 2.3768528891485317,
|
| 1057 |
+
"learning_rate": 7.532261550812585e-06,
|
| 1058 |
+
"loss": 0.5218,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.40085515766969537,
|
| 1063 |
+
"grad_norm": 2.525268737076288,
|
| 1064 |
+
"learning_rate": 7.491912764292764e-06,
|
| 1065 |
+
"loss": 0.5285,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4035275253874933,
|
| 1070 |
+
"grad_norm": 2.3135932700157515,
|
| 1071 |
+
"learning_rate": 7.451346907074245e-06,
|
| 1072 |
+
"loss": 0.5383,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.4061998931052913,
|
| 1077 |
+
"grad_norm": 2.2820011368190456,
|
| 1078 |
+
"learning_rate": 7.4105675128517456e-06,
|
| 1079 |
+
"loss": 0.5386,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.40887226082308925,
|
| 1084 |
+
"grad_norm": 2.6272733311462844,
|
| 1085 |
+
"learning_rate": 7.369578133921205e-06,
|
| 1086 |
+
"loss": 0.5403,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.41154462854088725,
|
| 1091 |
+
"grad_norm": 2.450459988951531,
|
| 1092 |
+
"learning_rate": 7.3283823408703466e-06,
|
| 1093 |
+
"loss": 0.5329,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4142169962586852,
|
| 1098 |
+
"grad_norm": 2.467816462124768,
|
| 1099 |
+
"learning_rate": 7.2869837222676445e-06,
|
| 1100 |
+
"loss": 0.5153,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4168893639764832,
|
| 1105 |
+
"grad_norm": 2.3951509707530962,
|
| 1106 |
+
"learning_rate": 7.245385884349716e-06,
|
| 1107 |
+
"loss": 0.5419,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4195617316942811,
|
| 1112 |
+
"grad_norm": 2.345171358544047,
|
| 1113 |
+
"learning_rate": 7.203592450707193e-06,
|
| 1114 |
+
"loss": 0.5221,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4222340994120791,
|
| 1119 |
+
"grad_norm": 2.37034917287765,
|
| 1120 |
+
"learning_rate": 7.161607061969061e-06,
|
| 1121 |
+
"loss": 0.5269,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.42490646712987706,
|
| 1126 |
+
"grad_norm": 2.3541847590463774,
|
| 1127 |
+
"learning_rate": 7.119433375485527e-06,
|
| 1128 |
+
"loss": 0.51,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.42757883484767506,
|
| 1133 |
+
"grad_norm": 2.4956945735597746,
|
| 1134 |
+
"learning_rate": 7.0770750650094335e-06,
|
| 1135 |
+
"loss": 0.529,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.430251202565473,
|
| 1140 |
+
"grad_norm": 2.228633956760255,
|
| 1141 |
+
"learning_rate": 7.034535820376225e-06,
|
| 1142 |
+
"loss": 0.5209,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.432923570283271,
|
| 1147 |
+
"grad_norm": 2.3668330274270675,
|
| 1148 |
+
"learning_rate": 6.991819347182536e-06,
|
| 1149 |
+
"loss": 0.5217,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.43559593800106894,
|
| 1154 |
+
"grad_norm": 2.440206726959897,
|
| 1155 |
+
"learning_rate": 6.948929366463397e-06,
|
| 1156 |
+
"loss": 0.5188,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.43826830571886694,
|
| 1161 |
+
"grad_norm": 2.281413045254601,
|
| 1162 |
+
"learning_rate": 6.9058696143680895e-06,
|
| 1163 |
+
"loss": 0.5048,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4409406734366649,
|
| 1168 |
+
"grad_norm": 2.2528776393400407,
|
| 1169 |
+
"learning_rate": 6.862643841834686e-06,
|
| 1170 |
+
"loss": 0.4963,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4436130411544629,
|
| 1175 |
+
"grad_norm": 2.298335166855743,
|
| 1176 |
+
"learning_rate": 6.8192558142633215e-06,
|
| 1177 |
+
"loss": 0.5167,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4462854088722608,
|
| 1182 |
+
"grad_norm": 2.3617346093820353,
|
| 1183 |
+
"learning_rate": 6.77570931118817e-06,
|
| 1184 |
+
"loss": 0.5038,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.4489577765900588,
|
| 1189 |
+
"grad_norm": 2.3822373788381577,
|
| 1190 |
+
"learning_rate": 6.732008125948223e-06,
|
| 1191 |
+
"loss": 0.5025,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.45163014430785675,
|
| 1196 |
+
"grad_norm": 2.5581676340917046,
|
| 1197 |
+
"learning_rate": 6.688156065356845e-06,
|
| 1198 |
+
"loss": 0.5054,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.45430251202565475,
|
| 1203 |
+
"grad_norm": 2.5460356769272354,
|
| 1204 |
+
"learning_rate": 6.644156949370162e-06,
|
| 1205 |
+
"loss": 0.517,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.4569748797434527,
|
| 1210 |
+
"grad_norm": 2.3347577676503635,
|
| 1211 |
+
"learning_rate": 6.600014610754306e-06,
|
| 1212 |
+
"loss": 0.503,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4596472474612507,
|
| 1217 |
+
"grad_norm": 2.457882693998114,
|
| 1218 |
+
"learning_rate": 6.555732894751548e-06,
|
| 1219 |
+
"loss": 0.5051,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4623196151790486,
|
| 1224 |
+
"grad_norm": 2.616065387749319,
|
| 1225 |
+
"learning_rate": 6.511315658745323e-06,
|
| 1226 |
+
"loss": 0.501,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4649919828968466,
|
| 1231 |
+
"grad_norm": 2.348479251904561,
|
| 1232 |
+
"learning_rate": 6.466766771924231e-06,
|
| 1233 |
+
"loss": 0.5113,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.46766435061464456,
|
| 1238 |
+
"grad_norm": 2.4230859156724414,
|
| 1239 |
+
"learning_rate": 6.422090114944982e-06,
|
| 1240 |
+
"loss": 0.4974,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.47033671833244256,
|
| 1245 |
+
"grad_norm": 2.2514326581575785,
|
| 1246 |
+
"learning_rate": 6.377289579594355e-06,
|
| 1247 |
+
"loss": 0.4922,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4730090860502405,
|
| 1252 |
+
"grad_norm": 2.3642088954621685,
|
| 1253 |
+
"learning_rate": 6.332369068450175e-06,
|
| 1254 |
+
"loss": 0.5041,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4756814537680385,
|
| 1259 |
+
"grad_norm": 2.508894790656293,
|
| 1260 |
+
"learning_rate": 6.28733249454138e-06,
|
| 1261 |
+
"loss": 0.5047,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.47835382148583644,
|
| 1266 |
+
"grad_norm": 2.3371495180507122,
|
| 1267 |
+
"learning_rate": 6.242183781007132e-06,
|
| 1268 |
+
"loss": 0.5009,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.48102618920363444,
|
| 1273 |
+
"grad_norm": 2.3182291284740013,
|
| 1274 |
+
"learning_rate": 6.196926860755088e-06,
|
| 1275 |
+
"loss": 0.4884,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4836985569214324,
|
| 1280 |
+
"grad_norm": 2.436532921181038,
|
| 1281 |
+
"learning_rate": 6.151565676118805e-06,
|
| 1282 |
+
"loss": 0.4961,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4863709246392304,
|
| 1287 |
+
"grad_norm": 2.4599123658932647,
|
| 1288 |
+
"learning_rate": 6.106104178514309e-06,
|
| 1289 |
+
"loss": 0.5068,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4890432923570283,
|
| 1294 |
+
"grad_norm": 2.328143651976841,
|
| 1295 |
+
"learning_rate": 6.0605463280958995e-06,
|
| 1296 |
+
"loss": 0.4843,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4917156600748263,
|
| 1301 |
+
"grad_norm": 2.2758994279070808,
|
| 1302 |
+
"learning_rate": 6.014896093411181e-06,
|
| 1303 |
+
"loss": 0.4856,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.49438802779262425,
|
| 1308 |
+
"grad_norm": 2.3653301550613484,
|
| 1309 |
+
"learning_rate": 5.9691574510553505e-06,
|
| 1310 |
+
"loss": 0.483,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.49706039551042225,
|
| 1315 |
+
"grad_norm": 2.4152543932592248,
|
| 1316 |
+
"learning_rate": 5.923334385324809e-06,
|
| 1317 |
+
"loss": 0.4684,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.4997327632282202,
|
| 1322 |
+
"grad_norm": 2.304808572699325,
|
| 1323 |
+
"learning_rate": 5.877430887870081e-06,
|
| 1324 |
+
"loss": 0.4765,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5024051309460181,
|
| 1329 |
+
"grad_norm": 2.357394956162013,
|
| 1330 |
+
"learning_rate": 5.831450957348106e-06,
|
| 1331 |
+
"loss": 0.4848,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5050774986638161,
|
| 1336 |
+
"grad_norm": 2.31372123260529,
|
| 1337 |
+
"learning_rate": 5.7853985990739115e-06,
|
| 1338 |
+
"loss": 0.4801,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5077498663816141,
|
| 1343 |
+
"grad_norm": 2.3349204861070225,
|
| 1344 |
+
"learning_rate": 5.739277824671711e-06,
|
| 1345 |
+
"loss": 0.4679,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5104222340994121,
|
| 1350 |
+
"grad_norm": 2.365618083997351,
|
| 1351 |
+
"learning_rate": 5.693092651725457e-06,
|
| 1352 |
+
"loss": 0.4858,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.51309460181721,
|
| 1357 |
+
"grad_norm": 2.289899872529692,
|
| 1358 |
+
"learning_rate": 5.646847103428859e-06,
|
| 1359 |
+
"loss": 0.4733,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.515766969535008,
|
| 1364 |
+
"grad_norm": 2.4225787265464374,
|
| 1365 |
+
"learning_rate": 5.600545208234927e-06,
|
| 1366 |
+
"loss": 0.4693,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.518439337252806,
|
| 1371 |
+
"grad_norm": 2.409714687251355,
|
| 1372 |
+
"learning_rate": 5.5541909995050554e-06,
|
| 1373 |
+
"loss": 0.4768,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.521111704970604,
|
| 1378 |
+
"grad_norm": 2.438137208132051,
|
| 1379 |
+
"learning_rate": 5.507788515157677e-06,
|
| 1380 |
+
"loss": 0.4688,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5237840726884019,
|
| 1385 |
+
"grad_norm": 2.3126195307394086,
|
| 1386 |
+
"learning_rate": 5.46134179731651e-06,
|
| 1387 |
+
"loss": 0.4838,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5264564404061999,
|
| 1392 |
+
"grad_norm": 2.431507168156365,
|
| 1393 |
+
"learning_rate": 5.414854891958464e-06,
|
| 1394 |
+
"loss": 0.4633,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5291288081239979,
|
| 1399 |
+
"grad_norm": 2.2665425075110117,
|
| 1400 |
+
"learning_rate": 5.368331848561178e-06,
|
| 1401 |
+
"loss": 0.4665,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5318011758417959,
|
| 1406 |
+
"grad_norm": 2.1898447124315097,
|
| 1407 |
+
"learning_rate": 5.321776719750283e-06,
|
| 1408 |
+
"loss": 0.4574,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5344735435595938,
|
| 1413 |
+
"grad_norm": 2.3240831234078048,
|
| 1414 |
+
"learning_rate": 5.275193560946372e-06,
|
| 1415 |
+
"loss": 0.4762,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5371459112773918,
|
| 1420 |
+
"grad_norm": 2.5172989799788827,
|
| 1421 |
+
"learning_rate": 5.228586430011732e-06,
|
| 1422 |
+
"loss": 0.4608,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5398182789951897,
|
| 1427 |
+
"grad_norm": 2.4058254619350015,
|
| 1428 |
+
"learning_rate": 5.181959386896862e-06,
|
| 1429 |
+
"loss": 0.4697,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5424906467129877,
|
| 1434 |
+
"grad_norm": 2.393865052227799,
|
| 1435 |
+
"learning_rate": 5.135316493286818e-06,
|
| 1436 |
+
"loss": 0.4547,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5451630144307856,
|
| 1441 |
+
"grad_norm": 2.279248494233032,
|
| 1442 |
+
"learning_rate": 5.088661812247389e-06,
|
| 1443 |
+
"loss": 0.4568,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5478353821485836,
|
| 1448 |
+
"grad_norm": 2.304039661249158,
|
| 1449 |
+
"learning_rate": 5.041999407871168e-06,
|
| 1450 |
+
"loss": 0.4687,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5505077498663816,
|
| 1455 |
+
"grad_norm": 2.2364242864738078,
|
| 1456 |
+
"learning_rate": 4.995333344923531e-06,
|
| 1457 |
+
"loss": 0.4618,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5531801175841796,
|
| 1462 |
+
"grad_norm": 2.3385130367436093,
|
| 1463 |
+
"learning_rate": 4.948667688488552e-06,
|
| 1464 |
+
"loss": 0.4538,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5558524853019775,
|
| 1469 |
+
"grad_norm": 2.1973220897038828,
|
| 1470 |
+
"learning_rate": 4.9020065036148885e-06,
|
| 1471 |
+
"loss": 0.4596,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5585248530197755,
|
| 1476 |
+
"grad_norm": 2.339222114519132,
|
| 1477 |
+
"learning_rate": 4.85535385496169e-06,
|
| 1478 |
+
"loss": 0.4639,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5611972207375735,
|
| 1483 |
+
"grad_norm": 2.343134563122755,
|
| 1484 |
+
"learning_rate": 4.808713806444506e-06,
|
| 1485 |
+
"loss": 0.4465,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5638695884553715,
|
| 1490 |
+
"grad_norm": 2.1561902388155403,
|
| 1491 |
+
"learning_rate": 4.762090420881289e-06,
|
| 1492 |
+
"loss": 0.4629,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5665419561731694,
|
| 1497 |
+
"grad_norm": 2.4064924815064734,
|
| 1498 |
+
"learning_rate": 4.715487759638486e-06,
|
| 1499 |
+
"loss": 0.4464,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5692143238909674,
|
| 1504 |
+
"grad_norm": 2.267003038981367,
|
| 1505 |
+
"learning_rate": 4.66890988227724e-06,
|
| 1506 |
+
"loss": 0.4605,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5718866916087654,
|
| 1511 |
+
"grad_norm": 2.313840406512139,
|
| 1512 |
+
"learning_rate": 4.622360846199772e-06,
|
| 1513 |
+
"loss": 0.4516,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5745590593265634,
|
| 1518 |
+
"grad_norm": 2.291001157001646,
|
| 1519 |
+
"learning_rate": 4.575844706295938e-06,
|
| 1520 |
+
"loss": 0.4326,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5772314270443613,
|
| 1525 |
+
"grad_norm": 2.358917294812099,
|
| 1526 |
+
"learning_rate": 4.529365514590002e-06,
|
| 1527 |
+
"loss": 0.4323,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5799037947621593,
|
| 1532 |
+
"grad_norm": 2.4267897696158505,
|
| 1533 |
+
"learning_rate": 4.482927319887669e-06,
|
| 1534 |
+
"loss": 0.4475,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5825761624799572,
|
| 1539 |
+
"grad_norm": 2.344155005554764,
|
| 1540 |
+
"learning_rate": 4.436534167423395e-06,
|
| 1541 |
+
"loss": 0.4382,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5852485301977552,
|
| 1546 |
+
"grad_norm": 2.1861870199235125,
|
| 1547 |
+
"learning_rate": 4.390190098508001e-06,
|
| 1548 |
+
"loss": 0.4254,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5879208979155531,
|
| 1553 |
+
"grad_norm": 2.2411000170176716,
|
| 1554 |
+
"learning_rate": 4.343899150176635e-06,
|
| 1555 |
+
"loss": 0.4392,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5905932656333511,
|
| 1560 |
+
"grad_norm": 2.4188630340463066,
|
| 1561 |
+
"learning_rate": 4.2976653548371115e-06,
|
| 1562 |
+
"loss": 0.4374,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5932656333511491,
|
| 1567 |
+
"grad_norm": 2.26211287377474,
|
| 1568 |
+
"learning_rate": 4.251492739918641e-06,
|
| 1569 |
+
"loss": 0.4334,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5959380010689471,
|
| 1574 |
+
"grad_norm": 2.3942328827051083,
|
| 1575 |
+
"learning_rate": 4.205385327521002e-06,
|
| 1576 |
+
"loss": 0.4438,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.598610368786745,
|
| 1581 |
+
"grad_norm": 2.1138070718766526,
|
| 1582 |
+
"learning_rate": 4.159347134064177e-06,
|
| 1583 |
+
"loss": 0.427,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.601282736504543,
|
| 1588 |
+
"grad_norm": 2.3128400895172065,
|
| 1589 |
+
"learning_rate": 4.113382169938488e-06,
|
| 1590 |
+
"loss": 0.4386,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.603955104222341,
|
| 1595 |
+
"grad_norm": 2.3090402861553008,
|
| 1596 |
+
"learning_rate": 4.067494439155236e-06,
|
| 1597 |
+
"loss": 0.4446,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.606627471940139,
|
| 1602 |
+
"grad_norm": 2.38557316695186,
|
| 1603 |
+
"learning_rate": 4.021687938997923e-06,
|
| 1604 |
+
"loss": 0.4283,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6092998396579369,
|
| 1609 |
+
"grad_norm": 2.3834734986558086,
|
| 1610 |
+
"learning_rate": 3.975966659674048e-06,
|
| 1611 |
+
"loss": 0.4318,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6119722073757349,
|
| 1616 |
+
"grad_norm": 2.468805695724146,
|
| 1617 |
+
"learning_rate": 3.930334583967514e-06,
|
| 1618 |
+
"loss": 0.4319,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6146445750935329,
|
| 1623 |
+
"grad_norm": 2.2597303719856754,
|
| 1624 |
+
"learning_rate": 3.884795686891692e-06,
|
| 1625 |
+
"loss": 0.4192,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6173169428113309,
|
| 1630 |
+
"grad_norm": 2.220457602337495,
|
| 1631 |
+
"learning_rate": 3.839353935343156e-06,
|
| 1632 |
+
"loss": 0.4387,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6199893105291288,
|
| 1637 |
+
"grad_norm": 2.2115364704480553,
|
| 1638 |
+
"learning_rate": 3.794013287756125e-06,
|
| 1639 |
+
"loss": 0.4119,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6226616782469268,
|
| 1644 |
+
"grad_norm": 2.3325364212519215,
|
| 1645 |
+
"learning_rate": 3.748777693757646e-06,
|
| 1646 |
+
"loss": 0.4315,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6253340459647247,
|
| 1651 |
+
"grad_norm": 2.305852431544535,
|
| 1652 |
+
"learning_rate": 3.7036510938235394e-06,
|
| 1653 |
+
"loss": 0.4379,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6280064136825227,
|
| 1658 |
+
"grad_norm": 2.315692835056829,
|
| 1659 |
+
"learning_rate": 3.658637418935146e-06,
|
| 1660 |
+
"loss": 0.4214,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6306787814003206,
|
| 1665 |
+
"grad_norm": 2.2325154851397655,
|
| 1666 |
+
"learning_rate": 3.613740590236895e-06,
|
| 1667 |
+
"loss": 0.4232,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6333511491181186,
|
| 1672 |
+
"grad_norm": 2.346809809533785,
|
| 1673 |
+
"learning_rate": 3.56896451869474e-06,
|
| 1674 |
+
"loss": 0.4096,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6360235168359166,
|
| 1679 |
+
"grad_norm": 2.2155776617857343,
|
| 1680 |
+
"learning_rate": 3.524313104755468e-06,
|
| 1681 |
+
"loss": 0.4257,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6386958845537146,
|
| 1686 |
+
"grad_norm": 2.301115466602763,
|
| 1687 |
+
"learning_rate": 3.4797902380069305e-06,
|
| 1688 |
+
"loss": 0.4161,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6413682522715125,
|
| 1693 |
+
"grad_norm": 2.267239712852779,
|
| 1694 |
+
"learning_rate": 3.4353997968392295e-06,
|
| 1695 |
+
"loss": 0.4134,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6440406199893105,
|
| 1700 |
+
"grad_norm": 2.314495503471501,
|
| 1701 |
+
"learning_rate": 3.3911456481068613e-06,
|
| 1702 |
+
"loss": 0.4246,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6467129877071085,
|
| 1707 |
+
"grad_norm": 2.2423488636530333,
|
| 1708 |
+
"learning_rate": 3.3470316467918785e-06,
|
| 1709 |
+
"loss": 0.4019,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6493853554249065,
|
| 1714 |
+
"grad_norm": 2.350376857517849,
|
| 1715 |
+
"learning_rate": 3.3030616356680854e-06,
|
| 1716 |
+
"loss": 0.4085,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6520577231427044,
|
| 1721 |
+
"grad_norm": 2.17587476305692,
|
| 1722 |
+
"learning_rate": 3.2592394449662867e-06,
|
| 1723 |
+
"loss": 0.4053,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6547300908605024,
|
| 1728 |
+
"grad_norm": 2.19222340194859,
|
| 1729 |
+
"learning_rate": 3.2155688920406415e-06,
|
| 1730 |
+
"loss": 0.4085,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6574024585783004,
|
| 1735 |
+
"grad_norm": 2.3348158981976534,
|
| 1736 |
+
"learning_rate": 3.172053781036132e-06,
|
| 1737 |
+
"loss": 0.4088,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6600748262960984,
|
| 1742 |
+
"grad_norm": 2.2791255065148777,
|
| 1743 |
+
"learning_rate": 3.1286979025571817e-06,
|
| 1744 |
+
"loss": 0.4162,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6627471940138963,
|
| 1749 |
+
"grad_norm": 2.184701125254012,
|
| 1750 |
+
"learning_rate": 3.0855050333374574e-06,
|
| 1751 |
+
"loss": 0.4146,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6654195617316943,
|
| 1756 |
+
"grad_norm": 2.1462756307983004,
|
| 1757 |
+
"learning_rate": 3.042478935910881e-06,
|
| 1758 |
+
"loss": 0.4066,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6680919294494923,
|
| 1763 |
+
"grad_norm": 2.249540103185502,
|
| 1764 |
+
"learning_rate": 2.9996233582838686e-06,
|
| 1765 |
+
"loss": 0.4123,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6707642971672902,
|
| 1770 |
+
"grad_norm": 2.1365843130945543,
|
| 1771 |
+
"learning_rate": 2.956942033608843e-06,
|
| 1772 |
+
"loss": 0.3986,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6734366648850882,
|
| 1777 |
+
"grad_norm": 2.2244554395915035,
|
| 1778 |
+
"learning_rate": 2.914438679859046e-06,
|
| 1779 |
+
"loss": 0.4022,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6761090326028861,
|
| 1784 |
+
"grad_norm": 2.1473916197507403,
|
| 1785 |
+
"learning_rate": 2.8721169995046503e-06,
|
| 1786 |
+
"loss": 0.4198,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6787814003206841,
|
| 1791 |
+
"grad_norm": 2.2135391990804245,
|
| 1792 |
+
"learning_rate": 2.829980679190254e-06,
|
| 1793 |
+
"loss": 0.3989,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6814537680384821,
|
| 1798 |
+
"grad_norm": 2.283258519501861,
|
| 1799 |
+
"learning_rate": 2.788033389413729e-06,
|
| 1800 |
+
"loss": 0.4021,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6841261357562801,
|
| 1805 |
+
"grad_norm": 2.3358515645118736,
|
| 1806 |
+
"learning_rate": 2.7462787842064753e-06,
|
| 1807 |
+
"loss": 0.4121,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.686798503474078,
|
| 1812 |
+
"grad_norm": 2.314456542436578,
|
| 1813 |
+
"learning_rate": 2.7047205008151332e-06,
|
| 1814 |
+
"loss": 0.4009,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.689470871191876,
|
| 1819 |
+
"grad_norm": 2.1923477921909176,
|
| 1820 |
+
"learning_rate": 2.6633621593847387e-06,
|
| 1821 |
+
"loss": 0.4021,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.692143238909674,
|
| 1826 |
+
"grad_norm": 2.2966892672617463,
|
| 1827 |
+
"learning_rate": 2.6222073626433587e-06,
|
| 1828 |
+
"loss": 0.395,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.694815606627472,
|
| 1833 |
+
"grad_norm": 2.2799392689553,
|
| 1834 |
+
"learning_rate": 2.5812596955882756e-06,
|
| 1835 |
+
"loss": 0.3949,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6974879743452699,
|
| 1840 |
+
"grad_norm": 2.2131564006746123,
|
| 1841 |
+
"learning_rate": 2.540522725173692e-06,
|
| 1842 |
+
"loss": 0.4001,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7001603420630679,
|
| 1847 |
+
"grad_norm": 2.2620859852805197,
|
| 1848 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1849 |
+
"loss": 0.3967,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7028327097808659,
|
| 1854 |
+
"grad_norm": 2.0891899787268278,
|
| 1855 |
+
"learning_rate": 2.459695050004688e-06,
|
| 1856 |
+
"loss": 0.3837,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7055050774986639,
|
| 1861 |
+
"grad_norm": 2.256364852897407,
|
| 1862 |
+
"learning_rate": 2.4196113861548233e-06,
|
| 1863 |
+
"loss": 0.3826,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7081774452164618,
|
| 1868 |
+
"grad_norm": 2.087597061423158,
|
| 1869 |
+
"learning_rate": 2.379752500141222e-06,
|
| 1870 |
+
"loss": 0.3904,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7108498129342598,
|
| 1875 |
+
"grad_norm": 2.148214107255344,
|
| 1876 |
+
"learning_rate": 2.3401218640742894e-06,
|
| 1877 |
+
"loss": 0.3925,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7135221806520577,
|
| 1882 |
+
"grad_norm": 2.252603568294772,
|
| 1883 |
+
"learning_rate": 2.3007229301815643e-06,
|
| 1884 |
+
"loss": 0.3843,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7161945483698557,
|
| 1889 |
+
"grad_norm": 2.5431488127516837,
|
| 1890 |
+
"learning_rate": 2.2615591305069846e-06,
|
| 1891 |
+
"loss": 0.3899,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7188669160876536,
|
| 1896 |
+
"grad_norm": 2.14235725429711,
|
| 1897 |
+
"learning_rate": 2.2226338766119366e-06,
|
| 1898 |
+
"loss": 0.3892,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7215392838054516,
|
| 1903 |
+
"grad_norm": 2.397791424619344,
|
| 1904 |
+
"learning_rate": 2.1839505592780658e-06,
|
| 1905 |
+
"loss": 0.3968,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7242116515232496,
|
| 1910 |
+
"grad_norm": 2.244839587602645,
|
| 1911 |
+
"learning_rate": 2.145512548211902e-06,
|
| 1912 |
+
"loss": 0.3661,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7268840192410476,
|
| 1917 |
+
"grad_norm": 2.081746722492156,
|
| 1918 |
+
"learning_rate": 2.1073231917513336e-06,
|
| 1919 |
+
"loss": 0.3805,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7295563869588455,
|
| 1924 |
+
"grad_norm": 2.1548041790438806,
|
| 1925 |
+
"learning_rate": 2.069385816573928e-06,
|
| 1926 |
+
"loss": 0.3739,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7322287546766435,
|
| 1931 |
+
"grad_norm": 2.2617609726198387,
|
| 1932 |
+
"learning_rate": 2.0317037274071412e-06,
|
| 1933 |
+
"loss": 0.3797,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7349011223944415,
|
| 1938 |
+
"grad_norm": 2.2088291878484694,
|
| 1939 |
+
"learning_rate": 1.99428020674045e-06,
|
| 1940 |
+
"loss": 0.3678,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7375734901122395,
|
| 1945 |
+
"grad_norm": 2.1855267811178645,
|
| 1946 |
+
"learning_rate": 1.9571185145394117e-06,
|
| 1947 |
+
"loss": 0.3873,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7402458578300374,
|
| 1952 |
+
"grad_norm": 2.1932190918187278,
|
| 1953 |
+
"learning_rate": 1.9202218879616824e-06,
|
| 1954 |
+
"loss": 0.3976,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7429182255478354,
|
| 1959 |
+
"grad_norm": 2.1791679190224142,
|
| 1960 |
+
"learning_rate": 1.8835935410750372e-06,
|
| 1961 |
+
"loss": 0.37,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7455905932656334,
|
| 1966 |
+
"grad_norm": 2.2552371440070043,
|
| 1967 |
+
"learning_rate": 1.8472366645773892e-06,
|
| 1968 |
+
"loss": 0.3728,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7482629609834314,
|
| 1973 |
+
"grad_norm": 2.146094844771457,
|
| 1974 |
+
"learning_rate": 1.8111544255188402e-06,
|
| 1975 |
+
"loss": 0.3875,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7509353287012293,
|
| 1980 |
+
"grad_norm": 2.018056735648775,
|
| 1981 |
+
"learning_rate": 1.7753499670258106e-06,
|
| 1982 |
+
"loss": 0.3763,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7536076964190273,
|
| 1987 |
+
"grad_norm": 2.076212339323707,
|
| 1988 |
+
"learning_rate": 1.7398264080272371e-06,
|
| 1989 |
+
"loss": 0.3849,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7562800641368252,
|
| 1994 |
+
"grad_norm": 2.174554150416714,
|
| 1995 |
+
"learning_rate": 1.7045868429828745e-06,
|
| 1996 |
+
"loss": 0.3856,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7589524318546232,
|
| 2001 |
+
"grad_norm": 2.151663250597911,
|
| 2002 |
+
"learning_rate": 1.6696343416137495e-06,
|
| 2003 |
+
"loss": 0.3861,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7616247995724211,
|
| 2008 |
+
"grad_norm": 2.3360649402814984,
|
| 2009 |
+
"learning_rate": 1.6349719486347533e-06,
|
| 2010 |
+
"loss": 0.3821,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7642971672902191,
|
| 2015 |
+
"grad_norm": 2.0553410775677854,
|
| 2016 |
+
"learning_rate": 1.6006026834894068e-06,
|
| 2017 |
+
"loss": 0.3669,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7669695350080171,
|
| 2022 |
+
"grad_norm": 2.1657944021551616,
|
| 2023 |
+
"learning_rate": 1.5665295400868513e-06,
|
| 2024 |
+
"loss": 0.3813,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7696419027258151,
|
| 2029 |
+
"grad_norm": 2.3170759830967325,
|
| 2030 |
+
"learning_rate": 1.5327554865410415e-06,
|
| 2031 |
+
"loss": 0.3723,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.772314270443613,
|
| 2036 |
+
"grad_norm": 2.1126770007065683,
|
| 2037 |
+
"learning_rate": 1.499283464912188e-06,
|
| 2038 |
+
"loss": 0.3706,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.774986638161411,
|
| 2043 |
+
"grad_norm": 2.372914223040231,
|
| 2044 |
+
"learning_rate": 1.4661163909504855e-06,
|
| 2045 |
+
"loss": 0.3771,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.777659005879209,
|
| 2050 |
+
"grad_norm": 2.3328613136831127,
|
| 2051 |
+
"learning_rate": 1.4332571538421136e-06,
|
| 2052 |
+
"loss": 0.3674,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.780331373597007,
|
| 2057 |
+
"grad_norm": 2.140027640464256,
|
| 2058 |
+
"learning_rate": 1.4007086159575595e-06,
|
| 2059 |
+
"loss": 0.3755,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7830037413148049,
|
| 2064 |
+
"grad_norm": 2.02535020358291,
|
| 2065 |
+
"learning_rate": 1.3684736126022812e-06,
|
| 2066 |
+
"loss": 0.3614,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7856761090326029,
|
| 2071 |
+
"grad_norm": 2.234825491106363,
|
| 2072 |
+
"learning_rate": 1.3365549517697234e-06,
|
| 2073 |
+
"loss": 0.3727,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7883484767504009,
|
| 2078 |
+
"grad_norm": 2.127621543244066,
|
| 2079 |
+
"learning_rate": 1.3049554138967052e-06,
|
| 2080 |
+
"loss": 0.3668,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7910208444681989,
|
| 2085 |
+
"grad_norm": 1.9966970197186302,
|
| 2086 |
+
"learning_rate": 1.2736777516212267e-06,
|
| 2087 |
+
"loss": 0.3792,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7936932121859968,
|
| 2092 |
+
"grad_norm": 2.1994020174447764,
|
| 2093 |
+
"learning_rate": 1.2427246895426826e-06,
|
| 2094 |
+
"loss": 0.3632,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7963655799037948,
|
| 2099 |
+
"grad_norm": 2.325490564770392,
|
| 2100 |
+
"learning_rate": 1.2120989239845149e-06,
|
| 2101 |
+
"loss": 0.3784,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7990379476215927,
|
| 2106 |
+
"grad_norm": 2.0389410793019094,
|
| 2107 |
+
"learning_rate": 1.1818031227593491e-06,
|
| 2108 |
+
"loss": 0.3688,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8017103153393907,
|
| 2113 |
+
"grad_norm": 2.237406582848762,
|
| 2114 |
+
"learning_rate": 1.1518399249365924e-06,
|
| 2115 |
+
"loss": 0.3675,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8043826830571886,
|
| 2120 |
+
"grad_norm": 2.0566468097872934,
|
| 2121 |
+
"learning_rate": 1.1222119406125426e-06,
|
| 2122 |
+
"loss": 0.3585,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8070550507749866,
|
| 2127 |
+
"grad_norm": 2.205571320640559,
|
| 2128 |
+
"learning_rate": 1.0929217506830292e-06,
|
| 2129 |
+
"loss": 0.3647,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8097274184927846,
|
| 2134 |
+
"grad_norm": 2.1023679887919635,
|
| 2135 |
+
"learning_rate": 1.0639719066185867e-06,
|
| 2136 |
+
"loss": 0.3715,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8123997862105826,
|
| 2141 |
+
"grad_norm": 2.0717279006060316,
|
| 2142 |
+
"learning_rate": 1.0353649302421982e-06,
|
| 2143 |
+
"loss": 0.3591,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8150721539283805,
|
| 2148 |
+
"grad_norm": 2.088163234375785,
|
| 2149 |
+
"learning_rate": 1.007103313509617e-06,
|
| 2150 |
+
"loss": 0.3569,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8177445216461785,
|
| 2155 |
+
"grad_norm": 2.2753464325922796,
|
| 2156 |
+
"learning_rate": 9.791895182922911e-07,
|
| 2157 |
+
"loss": 0.3633,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8204168893639765,
|
| 2162 |
+
"grad_norm": 2.122716745185861,
|
| 2163 |
+
"learning_rate": 9.516259761629148e-07,
|
| 2164 |
+
"loss": 0.3652,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8230892570817745,
|
| 2169 |
+
"grad_norm": 2.047095013957556,
|
| 2170 |
+
"learning_rate": 9.244150881836117e-07,
|
| 2171 |
+
"loss": 0.36,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8257616247995724,
|
| 2176 |
+
"grad_norm": 1.944446706318155,
|
| 2177 |
+
"learning_rate": 8.975592246967713e-07,
|
| 2178 |
+
"loss": 0.3653,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8284339925173704,
|
| 2183 |
+
"grad_norm": 2.038292799471268,
|
| 2184 |
+
"learning_rate": 8.710607251185799e-07,
|
| 2185 |
+
"loss": 0.3579,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8311063602351684,
|
| 2190 |
+
"grad_norm": 2.111895969332577,
|
| 2191 |
+
"learning_rate": 8.449218977352281e-07,
|
| 2192 |
+
"loss": 0.3489,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8337787279529664,
|
| 2197 |
+
"grad_norm": 2.0806458377180546,
|
| 2198 |
+
"learning_rate": 8.191450195018313e-07,
|
| 2199 |
+
"loss": 0.3612,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8364510956707643,
|
| 2204 |
+
"grad_norm": 2.2095760944892677,
|
| 2205 |
+
"learning_rate": 7.937323358440935e-07,
|
| 2206 |
+
"loss": 0.368,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8391234633885623,
|
| 2211 |
+
"grad_norm": 2.009589157938905,
|
| 2212 |
+
"learning_rate": 7.686860604627022e-07,
|
| 2213 |
+
"loss": 0.3697,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8417958311063602,
|
| 2218 |
+
"grad_norm": 2.0176043666039076,
|
| 2219 |
+
"learning_rate": 7.440083751404902e-07,
|
| 2220 |
+
"loss": 0.3479,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8444681988241582,
|
| 2225 |
+
"grad_norm": 2.2398694013369362,
|
| 2226 |
+
"learning_rate": 7.197014295523879e-07,
|
| 2227 |
+
"loss": 0.3658,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8471405665419561,
|
| 2232 |
+
"grad_norm": 2.2248765967199224,
|
| 2233 |
+
"learning_rate": 6.957673410781617e-07,
|
| 2234 |
+
"loss": 0.3612,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8498129342597541,
|
| 2239 |
+
"grad_norm": 2.1687807466267004,
|
| 2240 |
+
"learning_rate": 6.722081946179631e-07,
|
| 2241 |
+
"loss": 0.3537,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8524853019775521,
|
| 2246 |
+
"grad_norm": 2.152399358341962,
|
| 2247 |
+
"learning_rate": 6.490260424107231e-07,
|
| 2248 |
+
"loss": 0.3529,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8551576696953501,
|
| 2253 |
+
"grad_norm": 2.071713282983931,
|
| 2254 |
+
"learning_rate": 6.262229038553752e-07,
|
| 2255 |
+
"loss": 0.3541,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.857830037413148,
|
| 2260 |
+
"grad_norm": 2.0120433825076542,
|
| 2261 |
+
"learning_rate": 6.038007653349437e-07,
|
| 2262 |
+
"loss": 0.3672,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.860502405130946,
|
| 2267 |
+
"grad_norm": 2.091504907657452,
|
| 2268 |
+
"learning_rate": 5.817615800435167e-07,
|
| 2269 |
+
"loss": 0.3475,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.863174772848744,
|
| 2274 |
+
"grad_norm": 2.0847773360899837,
|
| 2275 |
+
"learning_rate": 5.60107267816098e-07,
|
| 2276 |
+
"loss": 0.3556,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.865847140566542,
|
| 2281 |
+
"grad_norm": 2.017553941427195,
|
| 2282 |
+
"learning_rate": 5.388397149613683e-07,
|
| 2283 |
+
"loss": 0.3713,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8685195082843399,
|
| 2288 |
+
"grad_norm": 2.3647834736903244,
|
| 2289 |
+
"learning_rate": 5.179607740973764e-07,
|
| 2290 |
+
"loss": 0.3631,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8711918760021379,
|
| 2295 |
+
"grad_norm": 2.041063738988079,
|
| 2296 |
+
"learning_rate": 4.974722639901503e-07,
|
| 2297 |
+
"loss": 0.3573,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8738642437199359,
|
| 2302 |
+
"grad_norm": 1.9173830344630454,
|
| 2303 |
+
"learning_rate": 4.773759693952662e-07,
|
| 2304 |
+
"loss": 0.3483,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8765366114377339,
|
| 2309 |
+
"grad_norm": 2.189591381428806,
|
| 2310 |
+
"learning_rate": 4.576736409023813e-07,
|
| 2311 |
+
"loss": 0.3617,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8792089791555318,
|
| 2316 |
+
"grad_norm": 2.075882031385073,
|
| 2317 |
+
"learning_rate": 4.383669947827368e-07,
|
| 2318 |
+
"loss": 0.3558,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8818813468733298,
|
| 2323 |
+
"grad_norm": 1.9859526858374763,
|
| 2324 |
+
"learning_rate": 4.194577128396521e-07,
|
| 2325 |
+
"loss": 0.3531,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8845537145911277,
|
| 2330 |
+
"grad_norm": 2.169320189826052,
|
| 2331 |
+
"learning_rate": 4.009474422620269e-07,
|
| 2332 |
+
"loss": 0.3597,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8872260823089257,
|
| 2337 |
+
"grad_norm": 2.0308766276062156,
|
| 2338 |
+
"learning_rate": 3.828377954808538e-07,
|
| 2339 |
+
"loss": 0.3471,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8898984500267236,
|
| 2344 |
+
"grad_norm": 2.0198511100844097,
|
| 2345 |
+
"learning_rate": 3.651303500287534e-07,
|
| 2346 |
+
"loss": 0.3576,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8925708177445216,
|
| 2351 |
+
"grad_norm": 2.268439126603322,
|
| 2352 |
+
"learning_rate": 3.4782664840256387e-07,
|
| 2353 |
+
"loss": 0.3529,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8952431854623196,
|
| 2358 |
+
"grad_norm": 1.9847511752450961,
|
| 2359 |
+
"learning_rate": 3.3092819792896913e-07,
|
| 2360 |
+
"loss": 0.3451,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8979155531801176,
|
| 2365 |
+
"grad_norm": 2.1019998352962164,
|
| 2366 |
+
"learning_rate": 3.1443647063319425e-07,
|
| 2367 |
+
"loss": 0.3453,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9005879208979155,
|
| 2372 |
+
"grad_norm": 2.2350648983618453,
|
| 2373 |
+
"learning_rate": 2.9835290311078123e-07,
|
| 2374 |
+
"loss": 0.3627,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9032602886157135,
|
| 2379 |
+
"grad_norm": 1.9841704743012314,
|
| 2380 |
+
"learning_rate": 2.8267889640244516e-07,
|
| 2381 |
+
"loss": 0.3535,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9059326563335115,
|
| 2386 |
+
"grad_norm": 2.0356033672039544,
|
| 2387 |
+
"learning_rate": 2.6741581587202747e-07,
|
| 2388 |
+
"loss": 0.3442,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9086050240513095,
|
| 2393 |
+
"grad_norm": 2.1092832232487826,
|
| 2394 |
+
"learning_rate": 2.525649910875627e-07,
|
| 2395 |
+
"loss": 0.3509,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9112773917691074,
|
| 2400 |
+
"grad_norm": 2.131995008854108,
|
| 2401 |
+
"learning_rate": 2.3812771570545846e-07,
|
| 2402 |
+
"loss": 0.3511,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9139497594869054,
|
| 2407 |
+
"grad_norm": 2.129385304232537,
|
| 2408 |
+
"learning_rate": 2.2410524735780205e-07,
|
| 2409 |
+
"loss": 0.3451,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9166221272047034,
|
| 2414 |
+
"grad_norm": 2.0136786336582877,
|
| 2415 |
+
"learning_rate": 2.104988075428127e-07,
|
| 2416 |
+
"loss": 0.3525,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9192944949225014,
|
| 2421 |
+
"grad_norm": 2.2858760868766765,
|
| 2422 |
+
"learning_rate": 1.9730958151843282e-07,
|
| 2423 |
+
"loss": 0.3484,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9219668626402993,
|
| 2428 |
+
"grad_norm": 1.964639607321287,
|
| 2429 |
+
"learning_rate": 1.845387181990821e-07,
|
| 2430 |
+
"loss": 0.3441,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9246392303580973,
|
| 2435 |
+
"grad_norm": 1.9899066828394374,
|
| 2436 |
+
"learning_rate": 1.7218733005557707e-07,
|
| 2437 |
+
"loss": 0.3562,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9273115980758952,
|
| 2442 |
+
"grad_norm": 2.1880261263868483,
|
| 2443 |
+
"learning_rate": 1.6025649301821877e-07,
|
| 2444 |
+
"loss": 0.3596,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9299839657936932,
|
| 2449 |
+
"grad_norm": 2.107871540829437,
|
| 2450 |
+
"learning_rate": 1.4874724638307303e-07,
|
| 2451 |
+
"loss": 0.3464,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9326563335114911,
|
| 2456 |
+
"grad_norm": 2.1515927738306635,
|
| 2457 |
+
"learning_rate": 1.376605927214364e-07,
|
| 2458 |
+
"loss": 0.3391,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9353287012292891,
|
| 2463 |
+
"grad_norm": 1.9419157718954796,
|
| 2464 |
+
"learning_rate": 1.2699749779249926e-07,
|
| 2465 |
+
"loss": 0.3578,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9380010689470871,
|
| 2470 |
+
"grad_norm": 2.0515419127475285,
|
| 2471 |
+
"learning_rate": 1.1675889045922151e-07,
|
| 2472 |
+
"loss": 0.3468,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9406734366648851,
|
| 2477 |
+
"grad_norm": 1.8780773791377043,
|
| 2478 |
+
"learning_rate": 1.0694566260742001e-07,
|
| 2479 |
+
"loss": 0.354,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.943345804382683,
|
| 2484 |
+
"grad_norm": 2.156801824942324,
|
| 2485 |
+
"learning_rate": 9.755866906807188e-08,
|
| 2486 |
+
"loss": 0.3529,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.946018172100481,
|
| 2491 |
+
"grad_norm": 2.049338756194342,
|
| 2492 |
+
"learning_rate": 8.859872754285403e-08,
|
| 2493 |
+
"loss": 0.3551,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.948690539818279,
|
| 2498 |
+
"grad_norm": 1.8992043254542215,
|
| 2499 |
+
"learning_rate": 8.006661853291298e-08,
|
| 2500 |
+
"loss": 0.3615,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.951362907536077,
|
| 2505 |
+
"grad_norm": 1.945897478392813,
|
| 2506 |
+
"learning_rate": 7.196308527087192e-08,
|
| 2507 |
+
"loss": 0.3489,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9540352752538749,
|
| 2512 |
+
"grad_norm": 1.934834420924745,
|
| 2513 |
+
"learning_rate": 6.428883365609261e-08,
|
| 2514 |
+
"loss": 0.3485,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9567076429716729,
|
| 2519 |
+
"grad_norm": 2.278947830724746,
|
| 2520 |
+
"learning_rate": 5.704453219318118e-08,
|
| 2521 |
+
"loss": 0.3465,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9593800106894709,
|
| 2526 |
+
"grad_norm": 2.019238988360205,
|
| 2527 |
+
"learning_rate": 5.023081193375357e-08,
|
| 2528 |
+
"loss": 0.3483,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9620523784072689,
|
| 2533 |
+
"grad_norm": 2.077457220463081,
|
| 2534 |
+
"learning_rate": 4.384826642146844e-08,
|
| 2535 |
+
"loss": 0.3512,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9647247461250668,
|
| 2540 |
+
"grad_norm": 1.90275137904516,
|
| 2541 |
+
"learning_rate": 3.7897451640321326e-08,
|
| 2542 |
+
"loss": 0.3336,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9673971138428648,
|
| 2547 |
+
"grad_norm": 2.2013361831715756,
|
| 2548 |
+
"learning_rate": 3.2378885966211636e-08,
|
| 2549 |
+
"loss": 0.3578,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9700694815606627,
|
| 2554 |
+
"grad_norm": 1.798115382458221,
|
| 2555 |
+
"learning_rate": 2.7293050121788843e-08,
|
| 2556 |
+
"loss": 0.3356,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9727418492784607,
|
| 2561 |
+
"grad_norm": 2.204761981011997,
|
| 2562 |
+
"learning_rate": 2.264038713457706e-08,
|
| 2563 |
+
"loss": 0.3564,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9754142169962586,
|
| 2568 |
+
"grad_norm": 2.114746372544894,
|
| 2569 |
+
"learning_rate": 1.842130229838035e-08,
|
| 2570 |
+
"loss": 0.3449,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9780865847140566,
|
| 2575 |
+
"grad_norm": 2.2465786265310825,
|
| 2576 |
+
"learning_rate": 1.4636163137980441e-08,
|
| 2577 |
+
"loss": 0.3496,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9807589524318546,
|
| 2582 |
+
"grad_norm": 2.4079788695162128,
|
| 2583 |
+
"learning_rate": 1.1285299377118974e-08,
|
| 2584 |
+
"loss": 0.3567,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9834313201496526,
|
| 2589 |
+
"grad_norm": 1.8850228312648118,
|
| 2590 |
+
"learning_rate": 8.369002909777713e-09,
|
| 2591 |
+
"loss": 0.3492,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9861036878674505,
|
| 2596 |
+
"grad_norm": 2.1554910144187702,
|
| 2597 |
+
"learning_rate": 5.88752777474999e-09,
|
| 2598 |
+
"loss": 0.3491,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9887760555852485,
|
| 2603 |
+
"grad_norm": 2.0257545592692794,
|
| 2604 |
+
"learning_rate": 3.841090133511749e-09,
|
| 2605 |
+
"loss": 0.3442,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9914484233030465,
|
| 2610 |
+
"grad_norm": 2.169390747670639,
|
| 2611 |
+
"learning_rate": 2.229868251391598e-09,
|
| 2612 |
+
"loss": 0.3502,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9941207910208445,
|
| 2617 |
+
"grad_norm": 2.169432604667664,
|
| 2618 |
+
"learning_rate": 1.054002482043237e-09,
|
| 2619 |
+
"loss": 0.361,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9967931587386424,
|
| 2624 |
+
"grad_norm": 1.9430919338940775,
|
| 2625 |
+
"learning_rate": 3.1359525521801326e-10,
|
| 2626 |
+
"loss": 0.3459,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9994655264564404,
|
| 2631 |
+
"grad_norm": 1.988798803083866,
|
| 2632 |
+
"learning_rate": 8.711067840949661e-12,
|
| 2633 |
+
"loss": 0.3484,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 1.0,
|
| 2638 |
+
"eval_runtime": 3.393,
|
| 2639 |
+
"eval_samples_per_second": 2.947,
|
| 2640 |
+
"eval_steps_per_second": 0.884,
|
| 2641 |
+
"step": 1871
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 1.0,
|
| 2645 |
+
"step": 1871,
|
| 2646 |
+
"total_flos": 195874667888640.0,
|
| 2647 |
+
"train_loss": 0.5146575008465732,
|
| 2648 |
+
"train_runtime": 16798.2789,
|
| 2649 |
+
"train_samples_per_second": 1.782,
|
| 2650 |
+
"train_steps_per_second": 0.111
|
| 2651 |
+
}
|
| 2652 |
+
],
|
| 2653 |
+
"logging_steps": 5,
|
| 2654 |
+
"max_steps": 1871,
|
| 2655 |
+
"num_input_tokens_seen": 0,
|
| 2656 |
+
"num_train_epochs": 1,
|
| 2657 |
+
"save_steps": 100,
|
| 2658 |
+
"stateful_callbacks": {
|
| 2659 |
+
"TrainerControl": {
|
| 2660 |
+
"args": {
|
| 2661 |
+
"should_epoch_stop": false,
|
| 2662 |
+
"should_evaluate": false,
|
| 2663 |
+
"should_log": false,
|
| 2664 |
+
"should_save": true,
|
| 2665 |
+
"should_training_stop": true
|
| 2666 |
+
},
|
| 2667 |
+
"attributes": {}
|
| 2668 |
+
}
|
| 2669 |
+
},
|
| 2670 |
+
"total_flos": 195874667888640.0,
|
| 2671 |
+
"train_batch_size": 4,
|
| 2672 |
+
"trial_name": null,
|
| 2673 |
+
"trial_params": null
|
| 2674 |
+
}
|