Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Aug08_02-12-33_COE-CS-sv003/events.out.tfevents.1723083489.COE-CS-sv003.3073689.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1485 -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: prometheus-7b-direct-0.1p-rare-seed42-response
|
| 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 |
+
# prometheus-7b-direct-0.1p-rare-seed42-response
|
| 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: 0.3360
|
| 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: 2
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- gradient_accumulation_steps: 4
|
| 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 |
+
| 0.3291 | 0.9998 | 1024 | 0.3360 |
|
| 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": 0.9997559189650964,
|
| 3 |
+
"total_flos": 214352422502400.0,
|
| 4 |
+
"train_loss": 0.48904780531302094,
|
| 5 |
+
"train_runtime": 23248.9436,
|
| 6 |
+
"train_samples": 108847,
|
| 7 |
+
"train_samples_per_second": 1.409,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 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/Aug08_02-12-33_COE-CS-sv003/events.out.tfevents.1723083489.COE-CS-sv003.3073689.0
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 sha256:5128753a710a6e6d84b8e1c4169ababf6f659e27fd8a05dc469e852a75ae63a9
|
| 3 |
+
size 49052
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9997559189650964,
|
| 3 |
+
"total_flos": 214352422502400.0,
|
| 4 |
+
"train_loss": 0.48904780531302094,
|
| 5 |
+
"train_runtime": 23248.9436,
|
| 6 |
+
"train_samples": 108847,
|
| 7 |
+
"train_samples_per_second": 1.409,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1485 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9997559189650964,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1024,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000976324139614352,
|
| 13 |
+
"grad_norm": 22.493754488892705,
|
| 14 |
+
"learning_rate": 9.70873786407767e-08,
|
| 15 |
+
"loss": 1.3084,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00488162069807176,
|
| 20 |
+
"grad_norm": 21.407549210638884,
|
| 21 |
+
"learning_rate": 4.854368932038835e-07,
|
| 22 |
+
"loss": 1.3239,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.00976324139614352,
|
| 27 |
+
"grad_norm": 9.073471250768227,
|
| 28 |
+
"learning_rate": 9.70873786407767e-07,
|
| 29 |
+
"loss": 1.2075,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01464486209421528,
|
| 34 |
+
"grad_norm": 8.959633641742187,
|
| 35 |
+
"learning_rate": 1.4563106796116506e-06,
|
| 36 |
+
"loss": 1.0555,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01952648279228704,
|
| 41 |
+
"grad_norm": 2.9726378530361903,
|
| 42 |
+
"learning_rate": 1.941747572815534e-06,
|
| 43 |
+
"loss": 0.9134,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0244081034903588,
|
| 48 |
+
"grad_norm": 2.336512387637656,
|
| 49 |
+
"learning_rate": 2.427184466019418e-06,
|
| 50 |
+
"loss": 0.8712,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02928972418843056,
|
| 55 |
+
"grad_norm": 2.1980743279610633,
|
| 56 |
+
"learning_rate": 2.912621359223301e-06,
|
| 57 |
+
"loss": 0.8452,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.034171344886502315,
|
| 62 |
+
"grad_norm": 2.2947165951756108,
|
| 63 |
+
"learning_rate": 3.398058252427185e-06,
|
| 64 |
+
"loss": 0.8304,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03905296558457408,
|
| 69 |
+
"grad_norm": 2.231374010631092,
|
| 70 |
+
"learning_rate": 3.883495145631068e-06,
|
| 71 |
+
"loss": 0.8005,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.043934586282645835,
|
| 76 |
+
"grad_norm": 2.323172717277238,
|
| 77 |
+
"learning_rate": 4.368932038834952e-06,
|
| 78 |
+
"loss": 0.7954,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0488162069807176,
|
| 83 |
+
"grad_norm": 2.6284940732364697,
|
| 84 |
+
"learning_rate": 4.854368932038836e-06,
|
| 85 |
+
"loss": 0.7828,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.053697827678789355,
|
| 90 |
+
"grad_norm": 2.4706362764533383,
|
| 91 |
+
"learning_rate": 5.3398058252427185e-06,
|
| 92 |
+
"loss": 0.7649,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.05857944837686112,
|
| 97 |
+
"grad_norm": 2.5587525974220915,
|
| 98 |
+
"learning_rate": 5.825242718446602e-06,
|
| 99 |
+
"loss": 0.7448,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.06346106907493287,
|
| 104 |
+
"grad_norm": 2.367706031948518,
|
| 105 |
+
"learning_rate": 6.310679611650487e-06,
|
| 106 |
+
"loss": 0.7313,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06834268977300463,
|
| 111 |
+
"grad_norm": 2.3861668688704416,
|
| 112 |
+
"learning_rate": 6.79611650485437e-06,
|
| 113 |
+
"loss": 0.7285,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0732243104710764,
|
| 118 |
+
"grad_norm": 2.2781470198507665,
|
| 119 |
+
"learning_rate": 7.2815533980582534e-06,
|
| 120 |
+
"loss": 0.7107,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07810593116914816,
|
| 125 |
+
"grad_norm": 2.4811146424704855,
|
| 126 |
+
"learning_rate": 7.766990291262136e-06,
|
| 127 |
+
"loss": 0.7127,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.08298755186721991,
|
| 132 |
+
"grad_norm": 2.3108856194291945,
|
| 133 |
+
"learning_rate": 8.25242718446602e-06,
|
| 134 |
+
"loss": 0.703,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08786917256529167,
|
| 139 |
+
"grad_norm": 2.4623970274283877,
|
| 140 |
+
"learning_rate": 8.737864077669904e-06,
|
| 141 |
+
"loss": 0.693,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.09275079326336344,
|
| 146 |
+
"grad_norm": 2.2177593778006544,
|
| 147 |
+
"learning_rate": 9.223300970873788e-06,
|
| 148 |
+
"loss": 0.6974,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0976324139614352,
|
| 153 |
+
"grad_norm": 2.3758805201515765,
|
| 154 |
+
"learning_rate": 9.708737864077671e-06,
|
| 155 |
+
"loss": 0.6771,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.10251403465950695,
|
| 160 |
+
"grad_norm": 2.367648483120376,
|
| 161 |
+
"learning_rate": 9.999883646674445e-06,
|
| 162 |
+
"loss": 0.703,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.10739565535757871,
|
| 167 |
+
"grad_norm": 2.7148702269535145,
|
| 168 |
+
"learning_rate": 9.998574733951775e-06,
|
| 169 |
+
"loss": 0.6806,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.11227727605565048,
|
| 174 |
+
"grad_norm": 2.3763948535285357,
|
| 175 |
+
"learning_rate": 9.995811848851807e-06,
|
| 176 |
+
"loss": 0.6747,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.11715889675372224,
|
| 181 |
+
"grad_norm": 2.3041091870777564,
|
| 182 |
+
"learning_rate": 9.991595795035352e-06,
|
| 183 |
+
"loss": 0.663,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.12204051745179399,
|
| 188 |
+
"grad_norm": 2.1733479043188644,
|
| 189 |
+
"learning_rate": 9.985927798857143e-06,
|
| 190 |
+
"loss": 0.677,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.12692213814986575,
|
| 195 |
+
"grad_norm": 2.2480309788297284,
|
| 196 |
+
"learning_rate": 9.978809509009121e-06,
|
| 197 |
+
"loss": 0.6625,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.13180375884793752,
|
| 202 |
+
"grad_norm": 2.138615471864636,
|
| 203 |
+
"learning_rate": 9.970242996040865e-06,
|
| 204 |
+
"loss": 0.656,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.13668537954600926,
|
| 209 |
+
"grad_norm": 2.1218250053659853,
|
| 210 |
+
"learning_rate": 9.960230751757318e-06,
|
| 211 |
+
"loss": 0.6676,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.14156700024408103,
|
| 216 |
+
"grad_norm": 2.067260559588222,
|
| 217 |
+
"learning_rate": 9.948775688493974e-06,
|
| 218 |
+
"loss": 0.6511,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1464486209421528,
|
| 223 |
+
"grad_norm": 2.0246056917859048,
|
| 224 |
+
"learning_rate": 9.93588113826975e-06,
|
| 225 |
+
"loss": 0.6514,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.15133024164022454,
|
| 230 |
+
"grad_norm": 2.2415009500993635,
|
| 231 |
+
"learning_rate": 9.921550851817774e-06,
|
| 232 |
+
"loss": 0.6554,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.15621186233829631,
|
| 237 |
+
"grad_norm": 2.2315931042976627,
|
| 238 |
+
"learning_rate": 9.905788997494377e-06,
|
| 239 |
+
"loss": 0.6499,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.16109348303636808,
|
| 244 |
+
"grad_norm": 2.15809328665933,
|
| 245 |
+
"learning_rate": 9.888600160066627e-06,
|
| 246 |
+
"loss": 0.6473,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.16597510373443983,
|
| 251 |
+
"grad_norm": 2.0579191397149836,
|
| 252 |
+
"learning_rate": 9.869989339378706e-06,
|
| 253 |
+
"loss": 0.6369,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1708567244325116,
|
| 258 |
+
"grad_norm": 2.031129747279531,
|
| 259 |
+
"learning_rate": 9.849961948897582e-06,
|
| 260 |
+
"loss": 0.6365,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.17573834513058334,
|
| 265 |
+
"grad_norm": 2.3571773219128067,
|
| 266 |
+
"learning_rate": 9.828523814138344e-06,
|
| 267 |
+
"loss": 0.6256,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1806199658286551,
|
| 272 |
+
"grad_norm": 2.0680850663572503,
|
| 273 |
+
"learning_rate": 9.8056811709697e-06,
|
| 274 |
+
"loss": 0.6305,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.18550158652672688,
|
| 279 |
+
"grad_norm": 2.070854390313614,
|
| 280 |
+
"learning_rate": 9.781440663800099e-06,
|
| 281 |
+
"loss": 0.6289,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.19038320722479862,
|
| 286 |
+
"grad_norm": 1.9927467214888446,
|
| 287 |
+
"learning_rate": 9.755809343645021e-06,
|
| 288 |
+
"loss": 0.6097,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1952648279228704,
|
| 293 |
+
"grad_norm": 2.1885730975002544,
|
| 294 |
+
"learning_rate": 9.728794666076004e-06,
|
| 295 |
+
"loss": 0.6202,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.20014644862094216,
|
| 300 |
+
"grad_norm": 2.0241934863146747,
|
| 301 |
+
"learning_rate": 9.700404489051974e-06,
|
| 302 |
+
"loss": 0.6218,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2050280693190139,
|
| 307 |
+
"grad_norm": 2.0845209209163076,
|
| 308 |
+
"learning_rate": 9.670647070633554e-06,
|
| 309 |
+
"loss": 0.6195,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.20990969001708568,
|
| 314 |
+
"grad_norm": 1.9439387252135285,
|
| 315 |
+
"learning_rate": 9.639531066580979e-06,
|
| 316 |
+
"loss": 0.6206,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.21479131071515742,
|
| 321 |
+
"grad_norm": 2.0645626959829184,
|
| 322 |
+
"learning_rate": 9.607065527836324e-06,
|
| 323 |
+
"loss": 0.6121,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2196729314132292,
|
| 328 |
+
"grad_norm": 2.0720595201040926,
|
| 329 |
+
"learning_rate": 9.573259897890794e-06,
|
| 330 |
+
"loss": 0.6146,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.22455455211130096,
|
| 335 |
+
"grad_norm": 1.9623515996646703,
|
| 336 |
+
"learning_rate": 9.538124010037832e-06,
|
| 337 |
+
"loss": 0.5982,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2294361728093727,
|
| 342 |
+
"grad_norm": 2.010779905137125,
|
| 343 |
+
"learning_rate": 9.501668084512827e-06,
|
| 344 |
+
"loss": 0.6062,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.23431779350744447,
|
| 349 |
+
"grad_norm": 2.1305241902290226,
|
| 350 |
+
"learning_rate": 9.46390272552028e-06,
|
| 351 |
+
"loss": 0.5966,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.23919941420551624,
|
| 356 |
+
"grad_norm": 1.8945487010213031,
|
| 357 |
+
"learning_rate": 9.424838918149285e-06,
|
| 358 |
+
"loss": 0.6016,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.24408103490358798,
|
| 363 |
+
"grad_norm": 1.9720726992448023,
|
| 364 |
+
"learning_rate": 9.384488025178214e-06,
|
| 365 |
+
"loss": 0.5908,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.24896265560165975,
|
| 370 |
+
"grad_norm": 2.5727183425693663,
|
| 371 |
+
"learning_rate": 9.342861783769535e-06,
|
| 372 |
+
"loss": 0.5892,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2538442762997315,
|
| 377 |
+
"grad_norm": 2.321112395905857,
|
| 378 |
+
"learning_rate": 9.29997230205575e-06,
|
| 379 |
+
"loss": 0.6011,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2587258969978033,
|
| 384 |
+
"grad_norm": 2.2030364900806076,
|
| 385 |
+
"learning_rate": 9.2558320556174e-06,
|
| 386 |
+
"loss": 0.5821,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.26360751769587504,
|
| 391 |
+
"grad_norm": 1.9812931636150968,
|
| 392 |
+
"learning_rate": 9.210453883854204e-06,
|
| 393 |
+
"loss": 0.5732,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2684891383939468,
|
| 398 |
+
"grad_norm": 2.085078299483637,
|
| 399 |
+
"learning_rate": 9.163850986250375e-06,
|
| 400 |
+
"loss": 0.5748,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2733707590920185,
|
| 405 |
+
"grad_norm": 2.2601821261422415,
|
| 406 |
+
"learning_rate": 9.11603691853518e-06,
|
| 407 |
+
"loss": 0.5704,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2782523797900903,
|
| 412 |
+
"grad_norm": 2.3842333315576076,
|
| 413 |
+
"learning_rate": 9.067025588739889e-06,
|
| 414 |
+
"loss": 0.5691,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.28313400048816206,
|
| 419 |
+
"grad_norm": 2.2224957447478038,
|
| 420 |
+
"learning_rate": 9.016831253152244e-06,
|
| 421 |
+
"loss": 0.5651,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2880156211862338,
|
| 426 |
+
"grad_norm": 2.1614870694331993,
|
| 427 |
+
"learning_rate": 8.96546851216962e-06,
|
| 428 |
+
"loss": 0.5559,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2928972418843056,
|
| 433 |
+
"grad_norm": 2.005310823414414,
|
| 434 |
+
"learning_rate": 8.912952306052109e-06,
|
| 435 |
+
"loss": 0.5657,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.29777886258237735,
|
| 440 |
+
"grad_norm": 2.38204601184677,
|
| 441 |
+
"learning_rate": 8.859297910576732e-06,
|
| 442 |
+
"loss": 0.5568,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.3026604832804491,
|
| 447 |
+
"grad_norm": 2.1648970631056743,
|
| 448 |
+
"learning_rate": 8.804520932594061e-06,
|
| 449 |
+
"loss": 0.5579,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3075421039785209,
|
| 454 |
+
"grad_norm": 2.100265583460356,
|
| 455 |
+
"learning_rate": 8.748637305488537e-06,
|
| 456 |
+
"loss": 0.5551,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.31242372467659263,
|
| 461 |
+
"grad_norm": 2.0675326704721138,
|
| 462 |
+
"learning_rate": 8.691663284543812e-06,
|
| 463 |
+
"loss": 0.5469,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.31730534537466437,
|
| 468 |
+
"grad_norm": 2.267073420915594,
|
| 469 |
+
"learning_rate": 8.633615442214452e-06,
|
| 470 |
+
"loss": 0.5561,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.32218696607273617,
|
| 475 |
+
"grad_norm": 2.087971484130279,
|
| 476 |
+
"learning_rate": 8.574510663305388e-06,
|
| 477 |
+
"loss": 0.5546,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3270685867708079,
|
| 482 |
+
"grad_norm": 1.8815922539302756,
|
| 483 |
+
"learning_rate": 8.514366140060504e-06,
|
| 484 |
+
"loss": 0.545,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.33195020746887965,
|
| 489 |
+
"grad_norm": 2.081332065948261,
|
| 490 |
+
"learning_rate": 8.453199367161804e-06,
|
| 491 |
+
"loss": 0.5409,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.33683182816695145,
|
| 496 |
+
"grad_norm": 2.0447565241323145,
|
| 497 |
+
"learning_rate": 8.391028136640604e-06,
|
| 498 |
+
"loss": 0.5345,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.3417134488650232,
|
| 503 |
+
"grad_norm": 1.8534927309067777,
|
| 504 |
+
"learning_rate": 8.32787053270223e-06,
|
| 505 |
+
"loss": 0.5295,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.34659506956309494,
|
| 510 |
+
"grad_norm": 2.0218669755624696,
|
| 511 |
+
"learning_rate": 8.263744926465744e-06,
|
| 512 |
+
"loss": 0.5339,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3514766902611667,
|
| 517 |
+
"grad_norm": 2.015374192428264,
|
| 518 |
+
"learning_rate": 8.198669970620177e-06,
|
| 519 |
+
"loss": 0.5296,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3563583109592385,
|
| 524 |
+
"grad_norm": 1.8414197150680836,
|
| 525 |
+
"learning_rate": 8.13266459399891e-06,
|
| 526 |
+
"loss": 0.5279,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3612399316573102,
|
| 531 |
+
"grad_norm": 1.9511713212083668,
|
| 532 |
+
"learning_rate": 8.065747996073681e-06,
|
| 533 |
+
"loss": 0.5201,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.36612155235538196,
|
| 538 |
+
"grad_norm": 2.0198178510951035,
|
| 539 |
+
"learning_rate": 7.997939641369909e-06,
|
| 540 |
+
"loss": 0.5231,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.37100317305345376,
|
| 545 |
+
"grad_norm": 1.9974814551196278,
|
| 546 |
+
"learning_rate": 7.929259253804903e-06,
|
| 547 |
+
"loss": 0.5127,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3758847937515255,
|
| 552 |
+
"grad_norm": 1.9584094585419525,
|
| 553 |
+
"learning_rate": 7.859726810950606e-06,
|
| 554 |
+
"loss": 0.503,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.38076641444959725,
|
| 559 |
+
"grad_norm": 1.9555704803278644,
|
| 560 |
+
"learning_rate": 7.789362538222585e-06,
|
| 561 |
+
"loss": 0.5031,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.38564803514766904,
|
| 566 |
+
"grad_norm": 1.9287309322500008,
|
| 567 |
+
"learning_rate": 7.718186902996912e-06,
|
| 568 |
+
"loss": 0.5095,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3905296558457408,
|
| 573 |
+
"grad_norm": 2.0211386074482998,
|
| 574 |
+
"learning_rate": 7.646220608656662e-06,
|
| 575 |
+
"loss": 0.4967,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.39541127654381253,
|
| 580 |
+
"grad_norm": 2.039367090773858,
|
| 581 |
+
"learning_rate": 7.573484588569775e-06,
|
| 582 |
+
"loss": 0.5015,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4002928972418843,
|
| 587 |
+
"grad_norm": 1.9208073920098578,
|
| 588 |
+
"learning_rate": 7.500000000000001e-06,
|
| 589 |
+
"loss": 0.497,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.40517451793995607,
|
| 594 |
+
"grad_norm": 1.8719019363210703,
|
| 595 |
+
"learning_rate": 7.425788217952744e-06,
|
| 596 |
+
"loss": 0.4955,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.4100561386380278,
|
| 601 |
+
"grad_norm": 1.9360771987365168,
|
| 602 |
+
"learning_rate": 7.350870828957547e-06,
|
| 603 |
+
"loss": 0.4906,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4149377593360996,
|
| 608 |
+
"grad_norm": 1.887372218244023,
|
| 609 |
+
"learning_rate": 7.27526962478906e-06,
|
| 610 |
+
"loss": 0.4896,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.41981938003417135,
|
| 615 |
+
"grad_norm": 1.965135998139074,
|
| 616 |
+
"learning_rate": 7.1990065961283075e-06,
|
| 617 |
+
"loss": 0.494,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.4247010007322431,
|
| 622 |
+
"grad_norm": 1.936347943982164,
|
| 623 |
+
"learning_rate": 7.122103926166096e-06,
|
| 624 |
+
"loss": 0.474,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.42958262143031484,
|
| 629 |
+
"grad_norm": 1.8906249170878375,
|
| 630 |
+
"learning_rate": 7.044583984150425e-06,
|
| 631 |
+
"loss": 0.4932,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.43446424212838664,
|
| 636 |
+
"grad_norm": 1.919261654779276,
|
| 637 |
+
"learning_rate": 6.9664693188797776e-06,
|
| 638 |
+
"loss": 0.4812,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.4393458628264584,
|
| 643 |
+
"grad_norm": 1.8837975572074062,
|
| 644 |
+
"learning_rate": 6.887782652144186e-06,
|
| 645 |
+
"loss": 0.477,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4442274835245301,
|
| 650 |
+
"grad_norm": 2.0761696776456904,
|
| 651 |
+
"learning_rate": 6.808546872115976e-06,
|
| 652 |
+
"loss": 0.4835,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4491091042226019,
|
| 657 |
+
"grad_norm": 1.9034620515101144,
|
| 658 |
+
"learning_rate": 6.728785026692113e-06,
|
| 659 |
+
"loss": 0.4692,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.45399072492067366,
|
| 664 |
+
"grad_norm": 2.0736151869195667,
|
| 665 |
+
"learning_rate": 6.648520316790102e-06,
|
| 666 |
+
"loss": 0.4585,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.4588723456187454,
|
| 671 |
+
"grad_norm": 2.011123921684231,
|
| 672 |
+
"learning_rate": 6.567776089599339e-06,
|
| 673 |
+
"loss": 0.4803,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.4637539663168172,
|
| 678 |
+
"grad_norm": 1.96249455995112,
|
| 679 |
+
"learning_rate": 6.486575831789974e-06,
|
| 680 |
+
"loss": 0.4623,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.46863558701488894,
|
| 685 |
+
"grad_norm": 2.2467468199710017,
|
| 686 |
+
"learning_rate": 6.404943162681144e-06,
|
| 687 |
+
"loss": 0.4574,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4735172077129607,
|
| 692 |
+
"grad_norm": 2.121288144154755,
|
| 693 |
+
"learning_rate": 6.322901827370659e-06,
|
| 694 |
+
"loss": 0.4681,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4783988284110325,
|
| 699 |
+
"grad_norm": 2.131811769895187,
|
| 700 |
+
"learning_rate": 6.240475689828087e-06,
|
| 701 |
+
"loss": 0.4573,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4832804491091042,
|
| 706 |
+
"grad_norm": 1.9175130135112601,
|
| 707 |
+
"learning_rate": 6.1576887259532695e-06,
|
| 708 |
+
"loss": 0.4465,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.48816206980717597,
|
| 713 |
+
"grad_norm": 1.8070710350230668,
|
| 714 |
+
"learning_rate": 6.074565016602263e-06,
|
| 715 |
+
"loss": 0.4419,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.49304369050524777,
|
| 720 |
+
"grad_norm": 1.99048575433868,
|
| 721 |
+
"learning_rate": 5.991128740582774e-06,
|
| 722 |
+
"loss": 0.4504,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.4979253112033195,
|
| 727 |
+
"grad_norm": 1.9210921458678536,
|
| 728 |
+
"learning_rate": 5.907404167621087e-06,
|
| 729 |
+
"loss": 0.4406,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5028069319013913,
|
| 734 |
+
"grad_norm": 2.009882308013614,
|
| 735 |
+
"learning_rate": 5.823415651302545e-06,
|
| 736 |
+
"loss": 0.4458,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.507688552599463,
|
| 741 |
+
"grad_norm": 1.897100615945393,
|
| 742 |
+
"learning_rate": 5.739187621987649e-06,
|
| 743 |
+
"loss": 0.4379,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5125701732975347,
|
| 748 |
+
"grad_norm": 1.907463353966728,
|
| 749 |
+
"learning_rate": 5.654744579705815e-06,
|
| 750 |
+
"loss": 0.4392,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5174517939956066,
|
| 755 |
+
"grad_norm": 2.112475304318316,
|
| 756 |
+
"learning_rate": 5.570111087028868e-06,
|
| 757 |
+
"loss": 0.4309,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5223334146936783,
|
| 762 |
+
"grad_norm": 1.9618257078116872,
|
| 763 |
+
"learning_rate": 5.4853117619263496e-06,
|
| 764 |
+
"loss": 0.4294,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5272150353917501,
|
| 769 |
+
"grad_norm": 2.023435280859463,
|
| 770 |
+
"learning_rate": 5.4003712706047055e-06,
|
| 771 |
+
"loss": 0.4412,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5320966560898218,
|
| 776 |
+
"grad_norm": 2.1255050556087003,
|
| 777 |
+
"learning_rate": 5.315314320332438e-06,
|
| 778 |
+
"loss": 0.4319,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5369782767878936,
|
| 783 |
+
"grad_norm": 2.0850824266858967,
|
| 784 |
+
"learning_rate": 5.230165652253329e-06,
|
| 785 |
+
"loss": 0.4365,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5418598974859653,
|
| 790 |
+
"grad_norm": 1.9630523459005746,
|
| 791 |
+
"learning_rate": 5.144950034189798e-06,
|
| 792 |
+
"loss": 0.4286,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.546741518184037,
|
| 797 |
+
"grad_norm": 1.8594385781324942,
|
| 798 |
+
"learning_rate": 5.059692253438495e-06,
|
| 799 |
+
"loss": 0.4216,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5516231388821089,
|
| 804 |
+
"grad_norm": 1.9410469701406243,
|
| 805 |
+
"learning_rate": 4.97441710956025e-06,
|
| 806 |
+
"loss": 0.4146,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.5565047595801806,
|
| 811 |
+
"grad_norm": 1.8926730305562134,
|
| 812 |
+
"learning_rate": 4.8891494071664315e-06,
|
| 813 |
+
"loss": 0.4243,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5613863802782524,
|
| 818 |
+
"grad_norm": 1.9153480931779545,
|
| 819 |
+
"learning_rate": 4.803913948703845e-06,
|
| 820 |
+
"loss": 0.4193,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5662680009763241,
|
| 825 |
+
"grad_norm": 1.9039605462186175,
|
| 826 |
+
"learning_rate": 4.71873552724027e-06,
|
| 827 |
+
"loss": 0.4098,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5711496216743959,
|
| 832 |
+
"grad_norm": 1.7928530873219377,
|
| 833 |
+
"learning_rate": 4.633638919252712e-06,
|
| 834 |
+
"loss": 0.4132,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5760312423724676,
|
| 839 |
+
"grad_norm": 1.8310404534316238,
|
| 840 |
+
"learning_rate": 4.548648877420481e-06,
|
| 841 |
+
"loss": 0.4056,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5809128630705395,
|
| 846 |
+
"grad_norm": 1.887050893521495,
|
| 847 |
+
"learning_rate": 4.463790123425209e-06,
|
| 848 |
+
"loss": 0.414,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5857944837686112,
|
| 853 |
+
"grad_norm": 1.8879105533071967,
|
| 854 |
+
"learning_rate": 4.379087340759861e-06,
|
| 855 |
+
"loss": 0.4105,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.590676104466683,
|
| 860 |
+
"grad_norm": 1.8851096816683868,
|
| 861 |
+
"learning_rate": 4.294565167548866e-06,
|
| 862 |
+
"loss": 0.4097,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5955577251647547,
|
| 867 |
+
"grad_norm": 1.8909509373180151,
|
| 868 |
+
"learning_rate": 4.2102481893814504e-06,
|
| 869 |
+
"loss": 0.4038,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6004393458628264,
|
| 874 |
+
"grad_norm": 2.0267286135794564,
|
| 875 |
+
"learning_rate": 4.1261609321602406e-06,
|
| 876 |
+
"loss": 0.401,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6053209665608982,
|
| 881 |
+
"grad_norm": 1.8447881878140413,
|
| 882 |
+
"learning_rate": 4.042327854967231e-06,
|
| 883 |
+
"loss": 0.3998,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.61020258725897,
|
| 888 |
+
"grad_norm": 1.8206856992603886,
|
| 889 |
+
"learning_rate": 3.958773342949196e-06,
|
| 890 |
+
"loss": 0.3957,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.6150842079570418,
|
| 895 |
+
"grad_norm": 1.989687612307625,
|
| 896 |
+
"learning_rate": 3.875521700224598e-06,
|
| 897 |
+
"loss": 0.3897,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6199658286551135,
|
| 902 |
+
"grad_norm": 1.7577176528030285,
|
| 903 |
+
"learning_rate": 3.7925971428140655e-06,
|
| 904 |
+
"loss": 0.3974,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6248474493531853,
|
| 909 |
+
"grad_norm": 1.7992538376367235,
|
| 910 |
+
"learning_rate": 3.71002379159651e-06,
|
| 911 |
+
"loss": 0.3933,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.629729070051257,
|
| 916 |
+
"grad_norm": 1.838523059460591,
|
| 917 |
+
"learning_rate": 3.627825665292899e-06,
|
| 918 |
+
"loss": 0.3961,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6346106907493287,
|
| 923 |
+
"grad_norm": 1.881394659994862,
|
| 924 |
+
"learning_rate": 3.546026673479755e-06,
|
| 925 |
+
"loss": 0.3938,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6394923114474005,
|
| 930 |
+
"grad_norm": 1.879760457077884,
|
| 931 |
+
"learning_rate": 3.464650609634403e-06,
|
| 932 |
+
"loss": 0.3929,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6443739321454723,
|
| 937 |
+
"grad_norm": 1.9193591824291845,
|
| 938 |
+
"learning_rate": 3.383721144213985e-06,
|
| 939 |
+
"loss": 0.3869,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.6492555528435441,
|
| 944 |
+
"grad_norm": 2.0119373139170724,
|
| 945 |
+
"learning_rate": 3.3032618177702546e-06,
|
| 946 |
+
"loss": 0.3851,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.6541371735416158,
|
| 951 |
+
"grad_norm": 1.9053960768739056,
|
| 952 |
+
"learning_rate": 3.2232960341021703e-06,
|
| 953 |
+
"loss": 0.3874,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.6590187942396876,
|
| 958 |
+
"grad_norm": 1.8532461602692916,
|
| 959 |
+
"learning_rate": 3.1438470534482547e-06,
|
| 960 |
+
"loss": 0.3822,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6639004149377593,
|
| 965 |
+
"grad_norm": 1.9222824143537547,
|
| 966 |
+
"learning_rate": 3.064937985720717e-06,
|
| 967 |
+
"loss": 0.3737,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.668782035635831,
|
| 972 |
+
"grad_norm": 1.7942220113298164,
|
| 973 |
+
"learning_rate": 2.9865917837833025e-06,
|
| 974 |
+
"loss": 0.3868,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.6736636563339029,
|
| 979 |
+
"grad_norm": 1.7376810506201046,
|
| 980 |
+
"learning_rate": 2.9088312367748257e-06,
|
| 981 |
+
"loss": 0.3763,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.6785452770319746,
|
| 986 |
+
"grad_norm": 1.9201828792137692,
|
| 987 |
+
"learning_rate": 2.8316789634803223e-06,
|
| 988 |
+
"loss": 0.3865,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6834268977300464,
|
| 993 |
+
"grad_norm": 1.9845153126888537,
|
| 994 |
+
"learning_rate": 2.75515740575176e-06,
|
| 995 |
+
"loss": 0.372,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.6883085184281181,
|
| 1000 |
+
"grad_norm": 1.7986907948997257,
|
| 1001 |
+
"learning_rate": 2.6792888219802017e-06,
|
| 1002 |
+
"loss": 0.3727,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6931901391261899,
|
| 1007 |
+
"grad_norm": 1.8814905894540856,
|
| 1008 |
+
"learning_rate": 2.604095280621354e-06,
|
| 1009 |
+
"loss": 0.3719,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6980717598242616,
|
| 1014 |
+
"grad_norm": 1.8333636960019017,
|
| 1015 |
+
"learning_rate": 2.529598653776349e-06,
|
| 1016 |
+
"loss": 0.3825,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7029533805223334,
|
| 1021 |
+
"grad_norm": 1.973156240281083,
|
| 1022 |
+
"learning_rate": 2.4558206108296394e-06,
|
| 1023 |
+
"loss": 0.3666,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7078350012204052,
|
| 1028 |
+
"grad_norm": 1.8104849203150244,
|
| 1029 |
+
"learning_rate": 2.3827826121458713e-06,
|
| 1030 |
+
"loss": 0.3681,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.712716621918477,
|
| 1035 |
+
"grad_norm": 1.7648438350645508,
|
| 1036 |
+
"learning_rate": 2.3105059028275467e-06,
|
| 1037 |
+
"loss": 0.3604,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.7175982426165487,
|
| 1042 |
+
"grad_norm": 1.7085215152593554,
|
| 1043 |
+
"learning_rate": 2.2390115065352974e-06,
|
| 1044 |
+
"loss": 0.3599,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7224798633146204,
|
| 1049 |
+
"grad_norm": 1.7691508841496486,
|
| 1050 |
+
"learning_rate": 2.16832021937259e-06,
|
| 1051 |
+
"loss": 0.361,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7273614840126922,
|
| 1056 |
+
"grad_norm": 1.7631951882565822,
|
| 1057 |
+
"learning_rate": 2.0984526038366005e-06,
|
| 1058 |
+
"loss": 0.3648,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7322431047107639,
|
| 1063 |
+
"grad_norm": 1.7905281564289197,
|
| 1064 |
+
"learning_rate": 2.0294289828370506e-06,
|
| 1065 |
+
"loss": 0.3593,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7371247254088358,
|
| 1070 |
+
"grad_norm": 1.7614811796241294,
|
| 1071 |
+
"learning_rate": 1.9612694337847334e-06,
|
| 1072 |
+
"loss": 0.3636,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7420063461069075,
|
| 1077 |
+
"grad_norm": 1.7427057794971907,
|
| 1078 |
+
"learning_rate": 1.8939937827514509e-06,
|
| 1079 |
+
"loss": 0.3514,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.7468879668049793,
|
| 1084 |
+
"grad_norm": 1.8108047484780578,
|
| 1085 |
+
"learning_rate": 1.8276215987030489e-06,
|
| 1086 |
+
"loss": 0.354,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.751769587503051,
|
| 1091 |
+
"grad_norm": 1.8460169409837044,
|
| 1092 |
+
"learning_rate": 1.7621721878072601e-06,
|
| 1093 |
+
"loss": 0.3536,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.7566512082011227,
|
| 1098 |
+
"grad_norm": 1.7192504649916136,
|
| 1099 |
+
"learning_rate": 1.6976645878179677e-06,
|
| 1100 |
+
"loss": 0.3523,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.7615328288991945,
|
| 1105 |
+
"grad_norm": 1.7659873554952885,
|
| 1106 |
+
"learning_rate": 1.6341175625375554e-06,
|
| 1107 |
+
"loss": 0.3556,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.7664144495972663,
|
| 1112 |
+
"grad_norm": 1.7642790853502184,
|
| 1113 |
+
"learning_rate": 1.5715495963589434e-06,
|
| 1114 |
+
"loss": 0.3505,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.7712960702953381,
|
| 1119 |
+
"grad_norm": 1.8214336145120926,
|
| 1120 |
+
"learning_rate": 1.509978888888894e-06,
|
| 1121 |
+
"loss": 0.3557,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.7761776909934098,
|
| 1126 |
+
"grad_norm": 1.693821479855387,
|
| 1127 |
+
"learning_rate": 1.4494233496541548e-06,
|
| 1128 |
+
"loss": 0.3533,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.7810593116914816,
|
| 1133 |
+
"grad_norm": 1.7741424308531102,
|
| 1134 |
+
"learning_rate": 1.3899005928919901e-06,
|
| 1135 |
+
"loss": 0.3436,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.7859409323895533,
|
| 1140 |
+
"grad_norm": 1.702567333467084,
|
| 1141 |
+
"learning_rate": 1.3314279324265922e-06,
|
| 1142 |
+
"loss": 0.3484,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.7908225530876251,
|
| 1147 |
+
"grad_norm": 1.7878487144245359,
|
| 1148 |
+
"learning_rate": 1.2740223766328813e-06,
|
| 1149 |
+
"loss": 0.3472,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.7957041737856968,
|
| 1154 |
+
"grad_norm": 1.8410165852894955,
|
| 1155 |
+
"learning_rate": 1.2177006234891548e-06,
|
| 1156 |
+
"loss": 0.3462,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.8005857944837687,
|
| 1161 |
+
"grad_norm": 1.5957050689906005,
|
| 1162 |
+
"learning_rate": 1.1624790557200255e-06,
|
| 1163 |
+
"loss": 0.3474,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8054674151818404,
|
| 1168 |
+
"grad_norm": 1.7840672497407395,
|
| 1169 |
+
"learning_rate": 1.1083737360310487e-06,
|
| 1170 |
+
"loss": 0.3366,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8103490358799121,
|
| 1175 |
+
"grad_norm": 1.7234306868229268,
|
| 1176 |
+
"learning_rate": 1.0554004024364573e-06,
|
| 1177 |
+
"loss": 0.3459,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8152306565779839,
|
| 1182 |
+
"grad_norm": 1.8763987301580518,
|
| 1183 |
+
"learning_rate": 1.0035744636813188e-06,
|
| 1184 |
+
"loss": 0.3399,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.8201122772760556,
|
| 1189 |
+
"grad_norm": 1.8526209211475768,
|
| 1190 |
+
"learning_rate": 9.529109947594834e-07,
|
| 1191 |
+
"loss": 0.3414,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.8249938979741274,
|
| 1196 |
+
"grad_norm": 1.6663345867535264,
|
| 1197 |
+
"learning_rate": 9.034247325286122e-07,
|
| 1198 |
+
"loss": 0.3443,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.8298755186721992,
|
| 1203 |
+
"grad_norm": 1.7676921955192966,
|
| 1204 |
+
"learning_rate": 8.551300714235494e-07,
|
| 1205 |
+
"loss": 0.3427,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.834757139370271,
|
| 1210 |
+
"grad_norm": 1.7531428909936075,
|
| 1211 |
+
"learning_rate": 8.080410592693183e-07,
|
| 1212 |
+
"loss": 0.335,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8396387600683427,
|
| 1217 |
+
"grad_norm": 1.8118502056267978,
|
| 1218 |
+
"learning_rate": 7.621713931949181e-07,
|
| 1219 |
+
"loss": 0.3484,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.8445203807664144,
|
| 1224 |
+
"grad_norm": 1.6928707382034682,
|
| 1225 |
+
"learning_rate": 7.175344156491432e-07,
|
| 1226 |
+
"loss": 0.3392,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.8494020014644862,
|
| 1231 |
+
"grad_norm": 1.7632689910362094,
|
| 1232 |
+
"learning_rate": 6.741431105195623e-07,
|
| 1233 |
+
"loss": 0.338,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.8542836221625579,
|
| 1238 |
+
"grad_norm": 1.7172615010190266,
|
| 1239 |
+
"learning_rate": 6.32010099355806e-07,
|
| 1240 |
+
"loss": 0.3386,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.8591652428606297,
|
| 1245 |
+
"grad_norm": 1.6578979212585563,
|
| 1246 |
+
"learning_rate": 5.911476376982333e-07,
|
| 1247 |
+
"loss": 0.3322,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.8640468635587015,
|
| 1252 |
+
"grad_norm": 1.6474380347956434,
|
| 1253 |
+
"learning_rate": 5.515676115130819e-07,
|
| 1254 |
+
"loss": 0.342,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.8689284842567733,
|
| 1259 |
+
"grad_norm": 1.6988649276270078,
|
| 1260 |
+
"learning_rate": 5.132815337351038e-07,
|
| 1261 |
+
"loss": 0.346,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.873810104954845,
|
| 1266 |
+
"grad_norm": 1.8306999314950443,
|
| 1267 |
+
"learning_rate": 4.763005409187155e-07,
|
| 1268 |
+
"loss": 0.3374,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.8786917256529168,
|
| 1273 |
+
"grad_norm": 1.631803186772389,
|
| 1274 |
+
"learning_rate": 4.406353899986221e-07,
|
| 1275 |
+
"loss": 0.3361,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.8835733463509885,
|
| 1280 |
+
"grad_norm": 1.709090450527214,
|
| 1281 |
+
"learning_rate": 4.06296455160875e-07,
|
| 1282 |
+
"loss": 0.3293,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.8884549670490602,
|
| 1287 |
+
"grad_norm": 1.583186301936809,
|
| 1288 |
+
"learning_rate": 3.732937248252472e-07,
|
| 1289 |
+
"loss": 0.3343,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.8933365877471321,
|
| 1294 |
+
"grad_norm": 1.6042173876546493,
|
| 1295 |
+
"learning_rate": 3.416367987398345e-07,
|
| 1296 |
+
"loss": 0.331,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.8982182084452038,
|
| 1301 |
+
"grad_norm": 1.6876494518887346,
|
| 1302 |
+
"learning_rate": 3.113348851887038e-07,
|
| 1303 |
+
"loss": 0.3292,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9030998291432756,
|
| 1308 |
+
"grad_norm": 1.6392793033144468,
|
| 1309 |
+
"learning_rate": 2.8239679831341126e-07,
|
| 1310 |
+
"loss": 0.3296,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.9079814498413473,
|
| 1315 |
+
"grad_norm": 1.6466791755980095,
|
| 1316 |
+
"learning_rate": 2.548309555491674e-07,
|
| 1317 |
+
"loss": 0.3348,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9128630705394191,
|
| 1322 |
+
"grad_norm": 1.6369494718827322,
|
| 1323 |
+
"learning_rate": 2.2864537517639618e-07,
|
| 1324 |
+
"loss": 0.3329,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.9177446912374908,
|
| 1329 |
+
"grad_norm": 1.7151132649607947,
|
| 1330 |
+
"learning_rate": 2.038476739883982e-07,
|
| 1331 |
+
"loss": 0.334,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.9226263119355627,
|
| 1336 |
+
"grad_norm": 1.7998544094471982,
|
| 1337 |
+
"learning_rate": 1.804450650757972e-07,
|
| 1338 |
+
"loss": 0.3366,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.9275079326336344,
|
| 1343 |
+
"grad_norm": 1.6113958011155078,
|
| 1344 |
+
"learning_rate": 1.5844435572841544e-07,
|
| 1345 |
+
"loss": 0.3199,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.9323895533317061,
|
| 1350 |
+
"grad_norm": 1.620629896748232,
|
| 1351 |
+
"learning_rate": 1.3785194545518965e-07,
|
| 1352 |
+
"loss": 0.3331,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.9372711740297779,
|
| 1357 |
+
"grad_norm": 1.570691984682264,
|
| 1358 |
+
"learning_rate": 1.1867382412269257e-07,
|
| 1359 |
+
"loss": 0.3266,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.9421527947278496,
|
| 1364 |
+
"grad_norm": 1.7541787796617982,
|
| 1365 |
+
"learning_rate": 1.0091557021282283e-07,
|
| 1366 |
+
"loss": 0.3349,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.9470344154259214,
|
| 1371 |
+
"grad_norm": 1.6021405868365253,
|
| 1372 |
+
"learning_rate": 8.458234920014685e-08,
|
| 1373 |
+
"loss": 0.3242,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.9519160361239931,
|
| 1378 |
+
"grad_norm": 1.6818332750612033,
|
| 1379 |
+
"learning_rate": 6.967891204937737e-08,
|
| 1380 |
+
"loss": 0.3272,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.956797656822065,
|
| 1385 |
+
"grad_norm": 1.6842336140953997,
|
| 1386 |
+
"learning_rate": 5.620959383343061e-08,
|
| 1387 |
+
"loss": 0.3335,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.9616792775201367,
|
| 1392 |
+
"grad_norm": 1.6333432398092111,
|
| 1393 |
+
"learning_rate": 4.417831247244819e-08,
|
| 1394 |
+
"loss": 0.3269,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.9665608982182085,
|
| 1399 |
+
"grad_norm": 1.5870095684725907,
|
| 1400 |
+
"learning_rate": 3.3588567594161625e-08,
|
| 1401 |
+
"loss": 0.3167,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.9714425189162802,
|
| 1406 |
+
"grad_norm": 1.645998653655925,
|
| 1407 |
+
"learning_rate": 2.4443439515933754e-08,
|
| 1408 |
+
"loss": 0.3345,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.9763241396143519,
|
| 1413 |
+
"grad_norm": 1.6627361867783679,
|
| 1414 |
+
"learning_rate": 1.6745588348758836e-08,
|
| 1415 |
+
"loss": 0.329,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.9812057603124237,
|
| 1420 |
+
"grad_norm": 1.6968847442052823,
|
| 1421 |
+
"learning_rate": 1.0497253223502035e-08,
|
| 1422 |
+
"loss": 0.328,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.9860873810104955,
|
| 1427 |
+
"grad_norm": 1.8141716265386114,
|
| 1428 |
+
"learning_rate": 5.700251639581544e-09,
|
| 1429 |
+
"loss": 0.3243,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.9909690017085673,
|
| 1434 |
+
"grad_norm": 1.6444141321089352,
|
| 1435 |
+
"learning_rate": 2.355978936303127e-09,
|
| 1436 |
+
"loss": 0.3277,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.995850622406639,
|
| 1441 |
+
"grad_norm": 1.5713108568101966,
|
| 1442 |
+
"learning_rate": 4.6540788698534735e-10,
|
| 1443 |
+
"loss": 0.3291,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.9997559189650964,
|
| 1448 |
+
"eval_loss": 0.3360166847705841,
|
| 1449 |
+
"eval_runtime": 96.8069,
|
| 1450 |
+
"eval_samples_per_second": 3.12,
|
| 1451 |
+
"eval_steps_per_second": 0.785,
|
| 1452 |
+
"step": 1024
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.9997559189650964,
|
| 1456 |
+
"step": 1024,
|
| 1457 |
+
"total_flos": 214352422502400.0,
|
| 1458 |
+
"train_loss": 0.48904780531302094,
|
| 1459 |
+
"train_runtime": 23248.9436,
|
| 1460 |
+
"train_samples_per_second": 1.409,
|
| 1461 |
+
"train_steps_per_second": 0.044
|
| 1462 |
+
}
|
| 1463 |
+
],
|
| 1464 |
+
"logging_steps": 5,
|
| 1465 |
+
"max_steps": 1024,
|
| 1466 |
+
"num_input_tokens_seen": 0,
|
| 1467 |
+
"num_train_epochs": 1,
|
| 1468 |
+
"save_steps": 100,
|
| 1469 |
+
"stateful_callbacks": {
|
| 1470 |
+
"TrainerControl": {
|
| 1471 |
+
"args": {
|
| 1472 |
+
"should_epoch_stop": false,
|
| 1473 |
+
"should_evaluate": false,
|
| 1474 |
+
"should_log": false,
|
| 1475 |
+
"should_save": true,
|
| 1476 |
+
"should_training_stop": true
|
| 1477 |
+
},
|
| 1478 |
+
"attributes": {}
|
| 1479 |
+
}
|
| 1480 |
+
},
|
| 1481 |
+
"total_flos": 214352422502400.0,
|
| 1482 |
+
"train_batch_size": 2,
|
| 1483 |
+
"trial_name": null,
|
| 1484 |
+
"trial_params": null
|
| 1485 |
+
}
|