Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep11_13-32-43_COE-CS-sv003/events.out.tfevents.1726061590.COE-CS-sv003.917532.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1359 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: feedback_p0.1_seed42_level3_raremix
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# feedback_p0.1_seed42_level3_raremix
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.2994
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 4
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- gradient_accumulation_steps: 2
|
| 48 |
+
- total_train_batch_size: 32
|
| 49 |
+
- total_eval_batch_size: 4
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 1
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 59 |
+
| 0.3562 | 0.9995 | 931 | 0.2994 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- Transformers 4.43.4
|
| 65 |
+
- Pytorch 2.3.1+cu121
|
| 66 |
+
- Datasets 2.19.1
|
| 67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9994632313472893,
|
| 3 |
+
"total_flos": 194827769610240.0,
|
| 4 |
+
"train_loss": 0.5084374696496478,
|
| 5 |
+
"train_runtime": 18282.9087,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.63,
|
| 8 |
+
"train_steps_per_second": 0.051
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.43.4"
|
| 6 |
+
}
|
runs/Sep11_13-32-43_COE-CS-sv003/events.out.tfevents.1726061590.COE-CS-sv003.917532.0
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 sha256:1a4ec8203c23c1d77040bdf336c13786b19151ed7fe28a9808172dd4b0d3e0b9
|
| 3 |
+
size 45210
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9994632313472893,
|
| 3 |
+
"total_flos": 194827769610240.0,
|
| 4 |
+
"train_loss": 0.5084374696496478,
|
| 5 |
+
"train_runtime": 18282.9087,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.63,
|
| 8 |
+
"train_steps_per_second": 0.051
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1359 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9994632313472893,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 931,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0010735373054213634,
|
| 13 |
+
"grad_norm": 22.789785654510972,
|
| 14 |
+
"learning_rate": 1.0638297872340426e-07,
|
| 15 |
+
"loss": 1.3534,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005367686527106817,
|
| 20 |
+
"grad_norm": 21.426953387742742,
|
| 21 |
+
"learning_rate": 5.319148936170213e-07,
|
| 22 |
+
"loss": 1.315,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010735373054213635,
|
| 27 |
+
"grad_norm": 8.681854043380088,
|
| 28 |
+
"learning_rate": 1.0638297872340427e-06,
|
| 29 |
+
"loss": 1.2029,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01610305958132045,
|
| 34 |
+
"grad_norm": 6.879600686982149,
|
| 35 |
+
"learning_rate": 1.595744680851064e-06,
|
| 36 |
+
"loss": 1.0391,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02147074610842727,
|
| 41 |
+
"grad_norm": 3.0275647149692793,
|
| 42 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 43 |
+
"loss": 0.9127,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.026838432635534086,
|
| 48 |
+
"grad_norm": 2.4868622752512706,
|
| 49 |
+
"learning_rate": 2.6595744680851065e-06,
|
| 50 |
+
"loss": 0.8614,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0322061191626409,
|
| 55 |
+
"grad_norm": 2.260349279940821,
|
| 56 |
+
"learning_rate": 3.191489361702128e-06,
|
| 57 |
+
"loss": 0.8364,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03757380568974772,
|
| 62 |
+
"grad_norm": 2.3883199937027952,
|
| 63 |
+
"learning_rate": 3.723404255319149e-06,
|
| 64 |
+
"loss": 0.8104,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04294149221685454,
|
| 69 |
+
"grad_norm": 2.2131867254029656,
|
| 70 |
+
"learning_rate": 4.255319148936171e-06,
|
| 71 |
+
"loss": 0.799,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04830917874396135,
|
| 76 |
+
"grad_norm": 2.192168023131458,
|
| 77 |
+
"learning_rate": 4.787234042553192e-06,
|
| 78 |
+
"loss": 0.7889,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.05367686527106817,
|
| 83 |
+
"grad_norm": 2.3212514172557133,
|
| 84 |
+
"learning_rate": 5.319148936170213e-06,
|
| 85 |
+
"loss": 0.7614,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.059044551798174985,
|
| 90 |
+
"grad_norm": 2.3347499759614307,
|
| 91 |
+
"learning_rate": 5.851063829787235e-06,
|
| 92 |
+
"loss": 0.7482,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0644122383252818,
|
| 97 |
+
"grad_norm": 2.285910235729559,
|
| 98 |
+
"learning_rate": 6.382978723404256e-06,
|
| 99 |
+
"loss": 0.7444,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.06977992485238863,
|
| 104 |
+
"grad_norm": 2.4172902264287255,
|
| 105 |
+
"learning_rate": 6.914893617021278e-06,
|
| 106 |
+
"loss": 0.7238,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07514761137949544,
|
| 111 |
+
"grad_norm": 2.3034733454455596,
|
| 112 |
+
"learning_rate": 7.446808510638298e-06,
|
| 113 |
+
"loss": 0.7231,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08051529790660225,
|
| 118 |
+
"grad_norm": 2.4091543127728756,
|
| 119 |
+
"learning_rate": 7.97872340425532e-06,
|
| 120 |
+
"loss": 0.7036,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08588298443370908,
|
| 125 |
+
"grad_norm": 2.2455472355866837,
|
| 126 |
+
"learning_rate": 8.510638297872341e-06,
|
| 127 |
+
"loss": 0.7073,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.09125067096081589,
|
| 132 |
+
"grad_norm": 2.4464200044770936,
|
| 133 |
+
"learning_rate": 9.042553191489362e-06,
|
| 134 |
+
"loss": 0.7078,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0966183574879227,
|
| 139 |
+
"grad_norm": 2.35752750364654,
|
| 140 |
+
"learning_rate": 9.574468085106385e-06,
|
| 141 |
+
"loss": 0.6963,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.10198604401502952,
|
| 146 |
+
"grad_norm": 2.1996004538409473,
|
| 147 |
+
"learning_rate": 9.999964780082996e-06,
|
| 148 |
+
"loss": 0.6979,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.10735373054213634,
|
| 153 |
+
"grad_norm": 2.203322438503073,
|
| 154 |
+
"learning_rate": 9.998732135085665e-06,
|
| 155 |
+
"loss": 0.6906,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.11272141706924316,
|
| 160 |
+
"grad_norm": 2.264863963599694,
|
| 161 |
+
"learning_rate": 9.995738990383743e-06,
|
| 162 |
+
"loss": 0.6866,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.11808910359634997,
|
| 167 |
+
"grad_norm": 2.2286418377556236,
|
| 168 |
+
"learning_rate": 9.990986400130607e-06,
|
| 169 |
+
"loss": 0.6782,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12345679012345678,
|
| 174 |
+
"grad_norm": 2.1240149641244175,
|
| 175 |
+
"learning_rate": 9.984476038137437e-06,
|
| 176 |
+
"loss": 0.6697,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.1288244766505636,
|
| 181 |
+
"grad_norm": 2.0873242470478854,
|
| 182 |
+
"learning_rate": 9.97621019728372e-06,
|
| 183 |
+
"loss": 0.6688,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13419216317767044,
|
| 188 |
+
"grad_norm": 2.4812498111624612,
|
| 189 |
+
"learning_rate": 9.966191788709716e-06,
|
| 190 |
+
"loss": 0.6631,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.13955984970477725,
|
| 195 |
+
"grad_norm": 2.343955721850536,
|
| 196 |
+
"learning_rate": 9.954424340791195e-06,
|
| 197 |
+
"loss": 0.6644,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.14492753623188406,
|
| 202 |
+
"grad_norm": 2.339967843619562,
|
| 203 |
+
"learning_rate": 9.940911997896774e-06,
|
| 204 |
+
"loss": 0.6642,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.15029522275899088,
|
| 209 |
+
"grad_norm": 2.2056817559033837,
|
| 210 |
+
"learning_rate": 9.925659518928316e-06,
|
| 211 |
+
"loss": 0.667,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1556629092860977,
|
| 216 |
+
"grad_norm": 1.9903249362862974,
|
| 217 |
+
"learning_rate": 9.908672275644898e-06,
|
| 218 |
+
"loss": 0.6444,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1610305958132045,
|
| 223 |
+
"grad_norm": 2.2039128348943744,
|
| 224 |
+
"learning_rate": 9.889956250770933e-06,
|
| 225 |
+
"loss": 0.6488,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.16639828234031132,
|
| 230 |
+
"grad_norm": 2.2078157899902706,
|
| 231 |
+
"learning_rate": 9.86951803588912e-06,
|
| 232 |
+
"loss": 0.6499,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.17176596886741816,
|
| 237 |
+
"grad_norm": 2.0109828014248774,
|
| 238 |
+
"learning_rate": 9.847364829118963e-06,
|
| 239 |
+
"loss": 0.6519,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.17713365539452497,
|
| 244 |
+
"grad_norm": 1.9664484718503512,
|
| 245 |
+
"learning_rate": 9.82350443258166e-06,
|
| 246 |
+
"loss": 0.6401,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.18250134192163178,
|
| 251 |
+
"grad_norm": 2.0667728485558308,
|
| 252 |
+
"learning_rate": 9.797945249652295e-06,
|
| 253 |
+
"loss": 0.6437,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1878690284487386,
|
| 258 |
+
"grad_norm": 2.0065876578800346,
|
| 259 |
+
"learning_rate": 9.770696282000245e-06,
|
| 260 |
+
"loss": 0.6346,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1932367149758454,
|
| 265 |
+
"grad_norm": 1.987617668179378,
|
| 266 |
+
"learning_rate": 9.741767126418898e-06,
|
| 267 |
+
"loss": 0.6356,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.19860440150295222,
|
| 272 |
+
"grad_norm": 2.04140704556592,
|
| 273 |
+
"learning_rate": 9.711167971445766e-06,
|
| 274 |
+
"loss": 0.6349,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.20397208803005903,
|
| 279 |
+
"grad_norm": 2.047535907923479,
|
| 280 |
+
"learning_rate": 9.67890959377418e-06,
|
| 281 |
+
"loss": 0.6298,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.20933977455716588,
|
| 286 |
+
"grad_norm": 1.9558416597814228,
|
| 287 |
+
"learning_rate": 9.645003354457872e-06,
|
| 288 |
+
"loss": 0.6349,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.2147074610842727,
|
| 293 |
+
"grad_norm": 1.8984622795525588,
|
| 294 |
+
"learning_rate": 9.60946119490972e-06,
|
| 295 |
+
"loss": 0.6142,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.2200751476113795,
|
| 300 |
+
"grad_norm": 2.3906003365767496,
|
| 301 |
+
"learning_rate": 9.57229563269612e-06,
|
| 302 |
+
"loss": 0.6222,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.22544283413848631,
|
| 307 |
+
"grad_norm": 1.8921023328729312,
|
| 308 |
+
"learning_rate": 9.533519757128426e-06,
|
| 309 |
+
"loss": 0.6186,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.23081052066559313,
|
| 314 |
+
"grad_norm": 1.9626099660140686,
|
| 315 |
+
"learning_rate": 9.49314722465304e-06,
|
| 316 |
+
"loss": 0.6258,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.23617820719269994,
|
| 321 |
+
"grad_norm": 2.0411979584264266,
|
| 322 |
+
"learning_rate": 9.451192254041759e-06,
|
| 323 |
+
"loss": 0.5979,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.24154589371980675,
|
| 328 |
+
"grad_norm": 2.1889414515575183,
|
| 329 |
+
"learning_rate": 9.407669621384073e-06,
|
| 330 |
+
"loss": 0.6083,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.24691358024691357,
|
| 335 |
+
"grad_norm": 2.085807605093685,
|
| 336 |
+
"learning_rate": 9.362594654883185e-06,
|
| 337 |
+
"loss": 0.5919,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2522812667740204,
|
| 342 |
+
"grad_norm": 2.0627204109631543,
|
| 343 |
+
"learning_rate": 9.31598322945759e-06,
|
| 344 |
+
"loss": 0.6034,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.2576489533011272,
|
| 349 |
+
"grad_norm": 1.9378180140246337,
|
| 350 |
+
"learning_rate": 9.267851761150092e-06,
|
| 351 |
+
"loss": 0.6038,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.26301663982823403,
|
| 356 |
+
"grad_norm": 3.1058701754069924,
|
| 357 |
+
"learning_rate": 9.218217201346251e-06,
|
| 358 |
+
"loss": 0.595,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.2683843263553409,
|
| 363 |
+
"grad_norm": 1.9991169619012592,
|
| 364 |
+
"learning_rate": 9.167097030804289e-06,
|
| 365 |
+
"loss": 0.5874,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.27375201288244766,
|
| 370 |
+
"grad_norm": 2.061863952952721,
|
| 371 |
+
"learning_rate": 9.114509253498554e-06,
|
| 372 |
+
"loss": 0.5969,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2791196994095545,
|
| 377 |
+
"grad_norm": 2.0907726187864926,
|
| 378 |
+
"learning_rate": 9.060472390278717e-06,
|
| 379 |
+
"loss": 0.5814,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2844873859366613,
|
| 384 |
+
"grad_norm": 2.7768778562710397,
|
| 385 |
+
"learning_rate": 9.005005472346923e-06,
|
| 386 |
+
"loss": 0.593,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2898550724637681,
|
| 391 |
+
"grad_norm": 2.0516068085641574,
|
| 392 |
+
"learning_rate": 8.948128034555212e-06,
|
| 393 |
+
"loss": 0.5849,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2952227589908749,
|
| 398 |
+
"grad_norm": 1.8867496167590758,
|
| 399 |
+
"learning_rate": 8.889860108525544e-06,
|
| 400 |
+
"loss": 0.5863,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.30059044551798175,
|
| 405 |
+
"grad_norm": 1.942421908800686,
|
| 406 |
+
"learning_rate": 8.83022221559489e-06,
|
| 407 |
+
"loss": 0.5672,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.3059581320450886,
|
| 412 |
+
"grad_norm": 2.0272368493119393,
|
| 413 |
+
"learning_rate": 8.76923535958783e-06,
|
| 414 |
+
"loss": 0.5623,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.3113258185721954,
|
| 419 |
+
"grad_norm": 2.1343642849157787,
|
| 420 |
+
"learning_rate": 8.706921019419237e-06,
|
| 421 |
+
"loss": 0.5732,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3166935050993022,
|
| 426 |
+
"grad_norm": 1.9323875772249934,
|
| 427 |
+
"learning_rate": 8.643301141529619e-06,
|
| 428 |
+
"loss": 0.5656,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.322061191626409,
|
| 433 |
+
"grad_norm": 2.037757830855184,
|
| 434 |
+
"learning_rate": 8.578398132155846e-06,
|
| 435 |
+
"loss": 0.5611,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.32742887815351585,
|
| 440 |
+
"grad_norm": 1.9398690166595873,
|
| 441 |
+
"learning_rate": 8.512234849439887e-06,
|
| 442 |
+
"loss": 0.5645,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.33279656468062263,
|
| 447 |
+
"grad_norm": 2.190033028948571,
|
| 448 |
+
"learning_rate": 8.444834595378434e-06,
|
| 449 |
+
"loss": 0.5622,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.33816425120772947,
|
| 454 |
+
"grad_norm": 2.221598674029084,
|
| 455 |
+
"learning_rate": 8.376221107616187e-06,
|
| 456 |
+
"loss": 0.546,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.3435319377348363,
|
| 461 |
+
"grad_norm": 2.021210342324902,
|
| 462 |
+
"learning_rate": 8.306418551085707e-06,
|
| 463 |
+
"loss": 0.5524,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.3488996242619431,
|
| 468 |
+
"grad_norm": 1.87069475903427,
|
| 469 |
+
"learning_rate": 8.23545150949679e-06,
|
| 470 |
+
"loss": 0.5447,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.35426731078904994,
|
| 475 |
+
"grad_norm": 1.880521268570416,
|
| 476 |
+
"learning_rate": 8.163344976678342e-06,
|
| 477 |
+
"loss": 0.5392,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3596349973161567,
|
| 482 |
+
"grad_norm": 2.14764020394395,
|
| 483 |
+
"learning_rate": 8.090124347775837e-06,
|
| 484 |
+
"loss": 0.547,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.36500268384326356,
|
| 489 |
+
"grad_norm": 1.9583788892413105,
|
| 490 |
+
"learning_rate": 8.0158154103074e-06,
|
| 491 |
+
"loss": 0.5487,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.37037037037037035,
|
| 496 |
+
"grad_norm": 2.053310730502184,
|
| 497 |
+
"learning_rate": 7.940444335081733e-06,
|
| 498 |
+
"loss": 0.5408,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.3757380568974772,
|
| 503 |
+
"grad_norm": 1.9003460290083882,
|
| 504 |
+
"learning_rate": 7.864037666981037e-06,
|
| 505 |
+
"loss": 0.543,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.38110574342458403,
|
| 510 |
+
"grad_norm": 1.9850745541543715,
|
| 511 |
+
"learning_rate": 7.786622315612182e-06,
|
| 512 |
+
"loss": 0.5405,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3864734299516908,
|
| 517 |
+
"grad_norm": 2.071517945939152,
|
| 518 |
+
"learning_rate": 7.708225545829446e-06,
|
| 519 |
+
"loss": 0.5331,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.39184111647879766,
|
| 524 |
+
"grad_norm": 1.9667341052567295,
|
| 525 |
+
"learning_rate": 7.6288749681321115e-06,
|
| 526 |
+
"loss": 0.5276,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.39720880300590444,
|
| 531 |
+
"grad_norm": 1.9124777050070318,
|
| 532 |
+
"learning_rate": 7.548598528940354e-06,
|
| 533 |
+
"loss": 0.5205,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.4025764895330113,
|
| 538 |
+
"grad_norm": 1.9540497432924613,
|
| 539 |
+
"learning_rate": 7.4674245007528135e-06,
|
| 540 |
+
"loss": 0.5293,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.40794417606011807,
|
| 545 |
+
"grad_norm": 2.0287532132914716,
|
| 546 |
+
"learning_rate": 7.385381472189321e-06,
|
| 547 |
+
"loss": 0.5183,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.4133118625872249,
|
| 552 |
+
"grad_norm": 2.021246551625832,
|
| 553 |
+
"learning_rate": 7.302498337922293e-06,
|
| 554 |
+
"loss": 0.5214,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.41867954911433175,
|
| 559 |
+
"grad_norm": 2.075086153282224,
|
| 560 |
+
"learning_rate": 7.218804288500343e-06,
|
| 561 |
+
"loss": 0.5051,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.42404723564143854,
|
| 566 |
+
"grad_norm": 1.9853201300706842,
|
| 567 |
+
"learning_rate": 7.134328800067684e-06,
|
| 568 |
+
"loss": 0.5143,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4294149221685454,
|
| 573 |
+
"grad_norm": 1.9075844135415194,
|
| 574 |
+
"learning_rate": 7.049101623982938e-06,
|
| 575 |
+
"loss": 0.5076,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.43478260869565216,
|
| 580 |
+
"grad_norm": 1.9963194922766025,
|
| 581 |
+
"learning_rate": 6.963152776341044e-06,
|
| 582 |
+
"loss": 0.5087,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.440150295222759,
|
| 587 |
+
"grad_norm": 2.0510112086805083,
|
| 588 |
+
"learning_rate": 6.876512527401897e-06,
|
| 589 |
+
"loss": 0.5092,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.4455179817498658,
|
| 594 |
+
"grad_norm": 1.8923092211852284,
|
| 595 |
+
"learning_rate": 6.789211390929497e-06,
|
| 596 |
+
"loss": 0.4966,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.45088566827697263,
|
| 601 |
+
"grad_norm": 2.015790172637043,
|
| 602 |
+
"learning_rate": 6.701280113445324e-06,
|
| 603 |
+
"loss": 0.4967,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4562533548040794,
|
| 608 |
+
"grad_norm": 1.9134734917899814,
|
| 609 |
+
"learning_rate": 6.6127496633997475e-06,
|
| 610 |
+
"loss": 0.5039,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.46162104133118625,
|
| 615 |
+
"grad_norm": 1.9388380453029543,
|
| 616 |
+
"learning_rate": 6.523651220265269e-06,
|
| 617 |
+
"loss": 0.4762,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.4669887278582931,
|
| 622 |
+
"grad_norm": 1.838971696143469,
|
| 623 |
+
"learning_rate": 6.434016163555452e-06,
|
| 624 |
+
"loss": 0.4894,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.4723564143853999,
|
| 629 |
+
"grad_norm": 1.9058911845628344,
|
| 630 |
+
"learning_rate": 6.343876061773385e-06,
|
| 631 |
+
"loss": 0.4858,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4777241009125067,
|
| 636 |
+
"grad_norm": 2.027908561619023,
|
| 637 |
+
"learning_rate": 6.2532626612936035e-06,
|
| 638 |
+
"loss": 0.4829,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.4830917874396135,
|
| 643 |
+
"grad_norm": 1.9646108142819012,
|
| 644 |
+
"learning_rate": 6.162207875181354e-06,
|
| 645 |
+
"loss": 0.4838,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.48845947396672035,
|
| 650 |
+
"grad_norm": 1.891035535041024,
|
| 651 |
+
"learning_rate": 6.070743771953157e-06,
|
| 652 |
+
"loss": 0.4815,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.49382716049382713,
|
| 657 |
+
"grad_norm": 1.9516352660221894,
|
| 658 |
+
"learning_rate": 5.978902564282616e-06,
|
| 659 |
+
"loss": 0.4811,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.499194847020934,
|
| 664 |
+
"grad_norm": 2.07837231336269,
|
| 665 |
+
"learning_rate": 5.886716597655472e-06,
|
| 666 |
+
"loss": 0.4676,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5045625335480408,
|
| 671 |
+
"grad_norm": 2.0338000671858443,
|
| 672 |
+
"learning_rate": 5.7942183389778536e-06,
|
| 673 |
+
"loss": 0.4751,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5099302200751477,
|
| 678 |
+
"grad_norm": 1.9544187569972904,
|
| 679 |
+
"learning_rate": 5.701440365141799e-06,
|
| 680 |
+
"loss": 0.4747,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5152979066022544,
|
| 685 |
+
"grad_norm": 2.0013443075194517,
|
| 686 |
+
"learning_rate": 5.608415351552014e-06,
|
| 687 |
+
"loss": 0.4737,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5206655931293612,
|
| 692 |
+
"grad_norm": 2.392764011448066,
|
| 693 |
+
"learning_rate": 5.515176060617945e-06,
|
| 694 |
+
"loss": 0.4661,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5260332796564681,
|
| 699 |
+
"grad_norm": 2.191779753390301,
|
| 700 |
+
"learning_rate": 5.421755330215223e-06,
|
| 701 |
+
"loss": 0.4633,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5314009661835749,
|
| 706 |
+
"grad_norm": 1.9807734795602283,
|
| 707 |
+
"learning_rate": 5.328186062120509e-06,
|
| 708 |
+
"loss": 0.4675,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5367686527106817,
|
| 713 |
+
"grad_norm": 1.8845793314459505,
|
| 714 |
+
"learning_rate": 5.23450121042383e-06,
|
| 715 |
+
"loss": 0.4485,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5421363392377885,
|
| 720 |
+
"grad_norm": 1.9504286157116426,
|
| 721 |
+
"learning_rate": 5.140733769922525e-06,
|
| 722 |
+
"loss": 0.4492,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.5475040257648953,
|
| 727 |
+
"grad_norm": 1.9251970055484684,
|
| 728 |
+
"learning_rate": 5.0469167645008245e-06,
|
| 729 |
+
"loss": 0.4537,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5528717122920022,
|
| 734 |
+
"grad_norm": 2.0900201692813503,
|
| 735 |
+
"learning_rate": 4.953083235499177e-06,
|
| 736 |
+
"loss": 0.4541,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.558239398819109,
|
| 741 |
+
"grad_norm": 1.914274695560957,
|
| 742 |
+
"learning_rate": 4.859266230077474e-06,
|
| 743 |
+
"loss": 0.4416,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5636070853462157,
|
| 748 |
+
"grad_norm": 2.0363927763394685,
|
| 749 |
+
"learning_rate": 4.7654987895761705e-06,
|
| 750 |
+
"loss": 0.4578,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5689747718733226,
|
| 755 |
+
"grad_norm": 1.906876750353807,
|
| 756 |
+
"learning_rate": 4.671813937879494e-06,
|
| 757 |
+
"loss": 0.4445,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5743424584004294,
|
| 762 |
+
"grad_norm": 1.9970794776707403,
|
| 763 |
+
"learning_rate": 4.5782446697847775e-06,
|
| 764 |
+
"loss": 0.4432,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5797101449275363,
|
| 769 |
+
"grad_norm": 1.9626727656883254,
|
| 770 |
+
"learning_rate": 4.484823939382056e-06,
|
| 771 |
+
"loss": 0.436,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5850778314546431,
|
| 776 |
+
"grad_norm": 1.936046454009477,
|
| 777 |
+
"learning_rate": 4.391584648447989e-06,
|
| 778 |
+
"loss": 0.4433,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5904455179817498,
|
| 783 |
+
"grad_norm": 1.919712775449399,
|
| 784 |
+
"learning_rate": 4.298559634858202e-06,
|
| 785 |
+
"loss": 0.4397,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5958132045088567,
|
| 790 |
+
"grad_norm": 1.882081042043698,
|
| 791 |
+
"learning_rate": 4.205781661022146e-06,
|
| 792 |
+
"loss": 0.4274,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.6011808910359635,
|
| 797 |
+
"grad_norm": 1.9706554854402294,
|
| 798 |
+
"learning_rate": 4.1132834023445304e-06,
|
| 799 |
+
"loss": 0.4358,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.6065485775630703,
|
| 804 |
+
"grad_norm": 1.8892003217320965,
|
| 805 |
+
"learning_rate": 4.021097435717386e-06,
|
| 806 |
+
"loss": 0.4203,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6119162640901772,
|
| 811 |
+
"grad_norm": 1.9387503523379348,
|
| 812 |
+
"learning_rate": 3.929256228046845e-06,
|
| 813 |
+
"loss": 0.4188,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6172839506172839,
|
| 818 |
+
"grad_norm": 1.9263320255464111,
|
| 819 |
+
"learning_rate": 3.837792124818647e-06,
|
| 820 |
+
"loss": 0.4197,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6226516371443908,
|
| 825 |
+
"grad_norm": 2.0679571176329636,
|
| 826 |
+
"learning_rate": 3.7467373387063973e-06,
|
| 827 |
+
"loss": 0.4104,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.6280193236714976,
|
| 832 |
+
"grad_norm": 1.88133441550826,
|
| 833 |
+
"learning_rate": 3.656123938226618e-06,
|
| 834 |
+
"loss": 0.4094,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6333870101986044,
|
| 839 |
+
"grad_norm": 1.8992502262176936,
|
| 840 |
+
"learning_rate": 3.5659838364445505e-06,
|
| 841 |
+
"loss": 0.4126,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6387546967257112,
|
| 846 |
+
"grad_norm": 1.8928527924901577,
|
| 847 |
+
"learning_rate": 3.476348779734732e-06,
|
| 848 |
+
"loss": 0.4245,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.644122383252818,
|
| 853 |
+
"grad_norm": 1.8811095797880513,
|
| 854 |
+
"learning_rate": 3.387250336600254e-06,
|
| 855 |
+
"loss": 0.4157,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6494900697799249,
|
| 860 |
+
"grad_norm": 2.1867786519525736,
|
| 861 |
+
"learning_rate": 3.298719886554677e-06,
|
| 862 |
+
"loss": 0.4029,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6548577563070317,
|
| 867 |
+
"grad_norm": 1.9359388771852108,
|
| 868 |
+
"learning_rate": 3.2107886090705035e-06,
|
| 869 |
+
"loss": 0.4073,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6602254428341385,
|
| 874 |
+
"grad_norm": 2.0767122537438834,
|
| 875 |
+
"learning_rate": 3.1234874725981045e-06,
|
| 876 |
+
"loss": 0.4105,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6655931293612453,
|
| 881 |
+
"grad_norm": 1.9568967104623627,
|
| 882 |
+
"learning_rate": 3.036847223658958e-06,
|
| 883 |
+
"loss": 0.4166,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6709608158883521,
|
| 888 |
+
"grad_norm": 1.8330033161355852,
|
| 889 |
+
"learning_rate": 2.950898376017064e-06,
|
| 890 |
+
"loss": 0.4007,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.6763285024154589,
|
| 895 |
+
"grad_norm": 1.9534116085299529,
|
| 896 |
+
"learning_rate": 2.865671199932318e-06,
|
| 897 |
+
"loss": 0.4094,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6816961889425658,
|
| 902 |
+
"grad_norm": 1.824576200330308,
|
| 903 |
+
"learning_rate": 2.781195711499658e-06,
|
| 904 |
+
"loss": 0.3975,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6870638754696726,
|
| 909 |
+
"grad_norm": 1.9397944016656081,
|
| 910 |
+
"learning_rate": 2.697501662077707e-06,
|
| 911 |
+
"loss": 0.3938,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6924315619967794,
|
| 916 |
+
"grad_norm": 1.9405147376762144,
|
| 917 |
+
"learning_rate": 2.6146185278106807e-06,
|
| 918 |
+
"loss": 0.3903,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6977992485238862,
|
| 923 |
+
"grad_norm": 1.8735166891783446,
|
| 924 |
+
"learning_rate": 2.5325754992471886e-06,
|
| 925 |
+
"loss": 0.3844,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.703166935050993,
|
| 930 |
+
"grad_norm": 1.8550788085650558,
|
| 931 |
+
"learning_rate": 2.4514014710596467e-06,
|
| 932 |
+
"loss": 0.3895,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.7085346215780999,
|
| 937 |
+
"grad_norm": 1.9768445053781223,
|
| 938 |
+
"learning_rate": 2.3711250318678906e-06,
|
| 939 |
+
"loss": 0.3976,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.7139023081052066,
|
| 944 |
+
"grad_norm": 1.8242433586161713,
|
| 945 |
+
"learning_rate": 2.2917744541705544e-06,
|
| 946 |
+
"loss": 0.3896,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7192699946323134,
|
| 951 |
+
"grad_norm": 1.8766511465002484,
|
| 952 |
+
"learning_rate": 2.2133776843878185e-06,
|
| 953 |
+
"loss": 0.3881,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7246376811594203,
|
| 958 |
+
"grad_norm": 1.8567022542334322,
|
| 959 |
+
"learning_rate": 2.1359623330189655e-06,
|
| 960 |
+
"loss": 0.3826,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7300053676865271,
|
| 965 |
+
"grad_norm": 1.8349926895031767,
|
| 966 |
+
"learning_rate": 2.059555664918268e-06,
|
| 967 |
+
"loss": 0.3888,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.735373054213634,
|
| 972 |
+
"grad_norm": 1.7933412905107806,
|
| 973 |
+
"learning_rate": 1.9841845896926022e-06,
|
| 974 |
+
"loss": 0.3861,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.7407407407407407,
|
| 979 |
+
"grad_norm": 1.8516583877916601,
|
| 980 |
+
"learning_rate": 1.9098756522241634e-06,
|
| 981 |
+
"loss": 0.3793,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7461084272678475,
|
| 986 |
+
"grad_norm": 1.8073808501807826,
|
| 987 |
+
"learning_rate": 1.8366550233216584e-06,
|
| 988 |
+
"loss": 0.3825,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7514761137949544,
|
| 993 |
+
"grad_norm": 2.0053953972861183,
|
| 994 |
+
"learning_rate": 1.7645484905032129e-06,
|
| 995 |
+
"loss": 0.3795,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.7568438003220612,
|
| 1000 |
+
"grad_norm": 1.7686721996317474,
|
| 1001 |
+
"learning_rate": 1.6935814489142937e-06,
|
| 1002 |
+
"loss": 0.3726,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7622114868491681,
|
| 1007 |
+
"grad_norm": 1.7884368236538568,
|
| 1008 |
+
"learning_rate": 1.6237788923838149e-06,
|
| 1009 |
+
"loss": 0.3681,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.7675791733762748,
|
| 1014 |
+
"grad_norm": 1.8578617970297808,
|
| 1015 |
+
"learning_rate": 1.555165404621567e-06,
|
| 1016 |
+
"loss": 0.37,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7729468599033816,
|
| 1021 |
+
"grad_norm": 1.8740014923246073,
|
| 1022 |
+
"learning_rate": 1.487765150560116e-06,
|
| 1023 |
+
"loss": 0.3736,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7783145464304885,
|
| 1028 |
+
"grad_norm": 1.9238343180583926,
|
| 1029 |
+
"learning_rate": 1.4216018678441558e-06,
|
| 1030 |
+
"loss": 0.376,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.7836822329575953,
|
| 1035 |
+
"grad_norm": 1.907761388210502,
|
| 1036 |
+
"learning_rate": 1.3566988584703817e-06,
|
| 1037 |
+
"loss": 0.3749,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.789049919484702,
|
| 1042 |
+
"grad_norm": 1.8440526734882297,
|
| 1043 |
+
"learning_rate": 1.293078980580766e-06,
|
| 1044 |
+
"loss": 0.3656,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7944176060118089,
|
| 1049 |
+
"grad_norm": 1.8162595531294223,
|
| 1050 |
+
"learning_rate": 1.2307646404121692e-06,
|
| 1051 |
+
"loss": 0.3635,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7997852925389157,
|
| 1056 |
+
"grad_norm": 1.8090188411422603,
|
| 1057 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1058 |
+
"loss": 0.3706,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.8051529790660226,
|
| 1063 |
+
"grad_norm": 1.8548359213105685,
|
| 1064 |
+
"learning_rate": 1.1101398914744565e-06,
|
| 1065 |
+
"loss": 0.3541,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.8105206655931294,
|
| 1070 |
+
"grad_norm": 1.7486366206671473,
|
| 1071 |
+
"learning_rate": 1.0518719654447896e-06,
|
| 1072 |
+
"loss": 0.3566,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8158883521202361,
|
| 1077 |
+
"grad_norm": 1.7220786134987836,
|
| 1078 |
+
"learning_rate": 9.949945276530782e-07,
|
| 1079 |
+
"loss": 0.3537,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.821256038647343,
|
| 1084 |
+
"grad_norm": 1.8563952966735642,
|
| 1085 |
+
"learning_rate": 9.395276097212841e-07,
|
| 1086 |
+
"loss": 0.377,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8266237251744498,
|
| 1091 |
+
"grad_norm": 1.8261975975646805,
|
| 1092 |
+
"learning_rate": 8.854907465014479e-07,
|
| 1093 |
+
"loss": 0.3564,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.8319914117015567,
|
| 1098 |
+
"grad_norm": 1.6700870770757226,
|
| 1099 |
+
"learning_rate": 8.329029691957124e-07,
|
| 1100 |
+
"loss": 0.3533,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8373590982286635,
|
| 1105 |
+
"grad_norm": 1.8714115240328053,
|
| 1106 |
+
"learning_rate": 7.817827986537508e-07,
|
| 1107 |
+
"loss": 0.3598,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8427267847557702,
|
| 1112 |
+
"grad_norm": 1.7265092801300406,
|
| 1113 |
+
"learning_rate": 7.321482388499096e-07,
|
| 1114 |
+
"loss": 0.3574,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8480944712828771,
|
| 1119 |
+
"grad_norm": 1.7192645034643335,
|
| 1120 |
+
"learning_rate": 6.840167705424106e-07,
|
| 1121 |
+
"loss": 0.3626,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8534621578099839,
|
| 1126 |
+
"grad_norm": 1.8788914839483877,
|
| 1127 |
+
"learning_rate": 6.374053451168166e-07,
|
| 1128 |
+
"loss": 0.3569,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8588298443370908,
|
| 1133 |
+
"grad_norm": 1.7223384963954047,
|
| 1134 |
+
"learning_rate": 5.92330378615929e-07,
|
| 1135 |
+
"loss": 0.3567,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.8641975308641975,
|
| 1140 |
+
"grad_norm": 1.84633207462375,
|
| 1141 |
+
"learning_rate": 5.488077459582425e-07,
|
| 1142 |
+
"loss": 0.3568,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8695652173913043,
|
| 1147 |
+
"grad_norm": 1.8156905089011117,
|
| 1148 |
+
"learning_rate": 5.068527753469604e-07,
|
| 1149 |
+
"loss": 0.3553,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8749329039184112,
|
| 1154 |
+
"grad_norm": 1.76653198854091,
|
| 1155 |
+
"learning_rate": 4.664802428715753e-07,
|
| 1156 |
+
"loss": 0.353,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.880300590445518,
|
| 1161 |
+
"grad_norm": 1.746120045538455,
|
| 1162 |
+
"learning_rate": 4.2770436730388166e-07,
|
| 1163 |
+
"loss": 0.3548,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8856682769726248,
|
| 1168 |
+
"grad_norm": 1.7810142936647677,
|
| 1169 |
+
"learning_rate": 3.9053880509028086e-07,
|
| 1170 |
+
"loss": 0.3546,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8910359634997316,
|
| 1175 |
+
"grad_norm": 1.7315549203684755,
|
| 1176 |
+
"learning_rate": 3.549966455421305e-07,
|
| 1177 |
+
"loss": 0.3555,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8964036500268384,
|
| 1182 |
+
"grad_norm": 1.9272563135988092,
|
| 1183 |
+
"learning_rate": 3.2109040622582186e-07,
|
| 1184 |
+
"loss": 0.3521,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.9017713365539453,
|
| 1189 |
+
"grad_norm": 1.7798988754363725,
|
| 1190 |
+
"learning_rate": 2.8883202855423676e-07,
|
| 1191 |
+
"loss": 0.3506,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.9071390230810521,
|
| 1196 |
+
"grad_norm": 1.7625732551828204,
|
| 1197 |
+
"learning_rate": 2.582328735811029e-07,
|
| 1198 |
+
"loss": 0.3534,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.9125067096081588,
|
| 1203 |
+
"grad_norm": 1.827775623748613,
|
| 1204 |
+
"learning_rate": 2.2930371799975593e-07,
|
| 1205 |
+
"loss": 0.3469,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.9178743961352657,
|
| 1210 |
+
"grad_norm": 1.7729865240896066,
|
| 1211 |
+
"learning_rate": 2.0205475034770606e-07,
|
| 1212 |
+
"loss": 0.3478,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.9232420826623725,
|
| 1217 |
+
"grad_norm": 1.7459386763584364,
|
| 1218 |
+
"learning_rate": 1.7649556741833995e-07,
|
| 1219 |
+
"loss": 0.3506,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9286097691894794,
|
| 1224 |
+
"grad_norm": 1.7638105650910396,
|
| 1225 |
+
"learning_rate": 1.5263517088103862e-07,
|
| 1226 |
+
"loss": 0.3501,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.9339774557165862,
|
| 1231 |
+
"grad_norm": 1.6634385736641513,
|
| 1232 |
+
"learning_rate": 1.304819641108801e-07,
|
| 1233 |
+
"loss": 0.3443,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.9393451422436929,
|
| 1238 |
+
"grad_norm": 1.7138313811653363,
|
| 1239 |
+
"learning_rate": 1.1004374922906846e-07,
|
| 1240 |
+
"loss": 0.3519,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.9447128287707998,
|
| 1245 |
+
"grad_norm": 1.7315956523188014,
|
| 1246 |
+
"learning_rate": 9.132772435510362e-08,
|
| 1247 |
+
"loss": 0.3518,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9500805152979066,
|
| 1252 |
+
"grad_norm": 1.83104419897327,
|
| 1253 |
+
"learning_rate": 7.434048107168523e-08,
|
| 1254 |
+
"loss": 0.3534,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.9554482018250134,
|
| 1259 |
+
"grad_norm": 1.8614069199591219,
|
| 1260 |
+
"learning_rate": 5.908800210322696e-08,
|
| 1261 |
+
"loss": 0.3529,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9608158883521203,
|
| 1266 |
+
"grad_norm": 1.791303700726749,
|
| 1267 |
+
"learning_rate": 4.55756592088058e-08,
|
| 1268 |
+
"loss": 0.3458,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.966183574879227,
|
| 1273 |
+
"grad_norm": 1.766280788851407,
|
| 1274 |
+
"learning_rate": 3.3808211290284886e-08,
|
| 1275 |
+
"loss": 0.3506,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9715512614063339,
|
| 1280 |
+
"grad_norm": 1.7183411966725215,
|
| 1281 |
+
"learning_rate": 2.378980271628195e-08,
|
| 1282 |
+
"loss": 0.3459,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.9769189479334407,
|
| 1287 |
+
"grad_norm": 1.6247222017291731,
|
| 1288 |
+
"learning_rate": 1.552396186256411e-08,
|
| 1289 |
+
"loss": 0.3469,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.9822866344605475,
|
| 1294 |
+
"grad_norm": 1.737776722109284,
|
| 1295 |
+
"learning_rate": 9.013599869394096e-09,
|
| 1296 |
+
"loss": 0.3432,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.9876543209876543,
|
| 1301 |
+
"grad_norm": 1.6970200431542546,
|
| 1302 |
+
"learning_rate": 4.261009616257638e-09,
|
| 1303 |
+
"loss": 0.3502,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9930220075147611,
|
| 1308 |
+
"grad_norm": 1.700070440972857,
|
| 1309 |
+
"learning_rate": 1.2678649143349485e-09,
|
| 1310 |
+
"loss": 0.3439,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.998389694041868,
|
| 1315 |
+
"grad_norm": 1.7621651797235254,
|
| 1316 |
+
"learning_rate": 3.5219917003948003e-11,
|
| 1317 |
+
"loss": 0.3562,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9994632313472893,
|
| 1322 |
+
"eval_loss": 0.2994081974029541,
|
| 1323 |
+
"eval_runtime": 1.1873,
|
| 1324 |
+
"eval_samples_per_second": 1.684,
|
| 1325 |
+
"eval_steps_per_second": 0.842,
|
| 1326 |
+
"step": 931
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.9994632313472893,
|
| 1330 |
+
"step": 931,
|
| 1331 |
+
"total_flos": 194827769610240.0,
|
| 1332 |
+
"train_loss": 0.5084374696496478,
|
| 1333 |
+
"train_runtime": 18282.9087,
|
| 1334 |
+
"train_samples_per_second": 1.63,
|
| 1335 |
+
"train_steps_per_second": 0.051
|
| 1336 |
+
}
|
| 1337 |
+
],
|
| 1338 |
+
"logging_steps": 5,
|
| 1339 |
+
"max_steps": 931,
|
| 1340 |
+
"num_input_tokens_seen": 0,
|
| 1341 |
+
"num_train_epochs": 1,
|
| 1342 |
+
"save_steps": 100,
|
| 1343 |
+
"stateful_callbacks": {
|
| 1344 |
+
"TrainerControl": {
|
| 1345 |
+
"args": {
|
| 1346 |
+
"should_epoch_stop": false,
|
| 1347 |
+
"should_evaluate": false,
|
| 1348 |
+
"should_log": false,
|
| 1349 |
+
"should_save": true,
|
| 1350 |
+
"should_training_stop": true
|
| 1351 |
+
},
|
| 1352 |
+
"attributes": {}
|
| 1353 |
+
}
|
| 1354 |
+
},
|
| 1355 |
+
"total_flos": 194827769610240.0,
|
| 1356 |
+
"train_batch_size": 4,
|
| 1357 |
+
"trial_name": null,
|
| 1358 |
+
"trial_params": null
|
| 1359 |
+
}
|