Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Aug07_04-44-38_COE-CS-sv003/events.out.tfevents.1723005899.COE-CS-sv003.2784240.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1492 -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-style-seed42
|
| 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-style-seed42
|
| 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: 0.3359
|
| 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 |
+
| 0.3467 | 0.9993 | 1025 | 0.3359 |
|
| 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": 0.9992688276870583,
|
| 3 |
+
"total_flos": 214561802158080.0,
|
| 4 |
+
"train_loss": 0.5001517156275307,
|
| 5 |
+
"train_runtime": 26987.6611,
|
| 6 |
+
"train_samples": 108847,
|
| 7 |
+
"train_samples_per_second": 1.216,
|
| 8 |
+
"train_steps_per_second": 0.038
|
| 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/Aug07_04-44-38_COE-CS-sv003/events.out.tfevents.1723005899.COE-CS-sv003.2784240.0
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:fee93a554dd8ee4afa405b2f6e2d83f1fc8525e5a8ad90f96274786af2b3720b
|
| 3 |
+
size 49231
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9992688276870583,
|
| 3 |
+
"total_flos": 214561802158080.0,
|
| 4 |
+
"train_loss": 0.5001517156275307,
|
| 5 |
+
"train_runtime": 26987.6611,
|
| 6 |
+
"train_samples": 108847,
|
| 7 |
+
"train_samples_per_second": 1.216,
|
| 8 |
+
"train_steps_per_second": 0.038
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1492 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9992688276870583,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1025,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0009748964172556666,
|
| 13 |
+
"grad_norm": 22.86059565076889,
|
| 14 |
+
"learning_rate": 9.70873786407767e-08,
|
| 15 |
+
"loss": 1.3065,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004874482086278333,
|
| 20 |
+
"grad_norm": 22.08339033526192,
|
| 21 |
+
"learning_rate": 4.854368932038835e-07,
|
| 22 |
+
"loss": 1.316,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.009748964172556666,
|
| 27 |
+
"grad_norm": 8.710899949749317,
|
| 28 |
+
"learning_rate": 9.70873786407767e-07,
|
| 29 |
+
"loss": 1.2194,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014623446258834999,
|
| 34 |
+
"grad_norm": 8.431065000460007,
|
| 35 |
+
"learning_rate": 1.4563106796116506e-06,
|
| 36 |
+
"loss": 1.0655,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01949792834511333,
|
| 41 |
+
"grad_norm": 3.0129778026962346,
|
| 42 |
+
"learning_rate": 1.941747572815534e-06,
|
| 43 |
+
"loss": 0.9273,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.024372410431391666,
|
| 48 |
+
"grad_norm": 2.4586920682855142,
|
| 49 |
+
"learning_rate": 2.427184466019418e-06,
|
| 50 |
+
"loss": 0.8841,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.029246892517669997,
|
| 55 |
+
"grad_norm": 2.1763561124357684,
|
| 56 |
+
"learning_rate": 2.912621359223301e-06,
|
| 57 |
+
"loss": 0.853,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03412137460394833,
|
| 62 |
+
"grad_norm": 2.2029840594866466,
|
| 63 |
+
"learning_rate": 3.398058252427185e-06,
|
| 64 |
+
"loss": 0.8341,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03899585669022666,
|
| 69 |
+
"grad_norm": 2.446000778730272,
|
| 70 |
+
"learning_rate": 3.883495145631068e-06,
|
| 71 |
+
"loss": 0.8118,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.043870338776504994,
|
| 76 |
+
"grad_norm": 2.451352004117067,
|
| 77 |
+
"learning_rate": 4.368932038834952e-06,
|
| 78 |
+
"loss": 0.7943,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04874482086278333,
|
| 83 |
+
"grad_norm": 2.4129113465126957,
|
| 84 |
+
"learning_rate": 4.854368932038836e-06,
|
| 85 |
+
"loss": 0.7768,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05361930294906166,
|
| 90 |
+
"grad_norm": 2.3127065430589298,
|
| 91 |
+
"learning_rate": 5.3398058252427185e-06,
|
| 92 |
+
"loss": 0.7778,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.058493785035339994,
|
| 97 |
+
"grad_norm": 2.405876637403882,
|
| 98 |
+
"learning_rate": 5.825242718446602e-06,
|
| 99 |
+
"loss": 0.7534,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.06336826712161833,
|
| 104 |
+
"grad_norm": 2.2639828354846387,
|
| 105 |
+
"learning_rate": 6.310679611650487e-06,
|
| 106 |
+
"loss": 0.7398,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06824274920789666,
|
| 111 |
+
"grad_norm": 2.3243442460154657,
|
| 112 |
+
"learning_rate": 6.79611650485437e-06,
|
| 113 |
+
"loss": 0.7261,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.073117231294175,
|
| 118 |
+
"grad_norm": 2.4227745775819836,
|
| 119 |
+
"learning_rate": 7.2815533980582534e-06,
|
| 120 |
+
"loss": 0.7168,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07799171338045333,
|
| 125 |
+
"grad_norm": 4.25071801960126,
|
| 126 |
+
"learning_rate": 7.766990291262136e-06,
|
| 127 |
+
"loss": 0.71,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.08286619546673166,
|
| 132 |
+
"grad_norm": 2.148825747496477,
|
| 133 |
+
"learning_rate": 8.25242718446602e-06,
|
| 134 |
+
"loss": 0.7074,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08774067755300999,
|
| 139 |
+
"grad_norm": 2.380339632272055,
|
| 140 |
+
"learning_rate": 8.737864077669904e-06,
|
| 141 |
+
"loss": 0.7032,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.09261515963928832,
|
| 146 |
+
"grad_norm": 2.3544801176511916,
|
| 147 |
+
"learning_rate": 9.223300970873788e-06,
|
| 148 |
+
"loss": 0.7,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.09748964172556666,
|
| 153 |
+
"grad_norm": 2.1648774285785013,
|
| 154 |
+
"learning_rate": 9.708737864077671e-06,
|
| 155 |
+
"loss": 0.6812,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.102364123811845,
|
| 160 |
+
"grad_norm": 2.103759384635341,
|
| 161 |
+
"learning_rate": 9.999883898929927e-06,
|
| 162 |
+
"loss": 0.6822,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.10723860589812333,
|
| 167 |
+
"grad_norm": 2.1317773929603576,
|
| 168 |
+
"learning_rate": 9.998577823812066e-06,
|
| 169 |
+
"loss": 0.6845,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.11211308798440166,
|
| 174 |
+
"grad_norm": 2.5038388167151924,
|
| 175 |
+
"learning_rate": 9.995820927586548e-06,
|
| 176 |
+
"loss": 0.6799,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.11698757007067999,
|
| 181 |
+
"grad_norm": 2.2890913021002772,
|
| 182 |
+
"learning_rate": 9.99161401043362e-06,
|
| 183 |
+
"loss": 0.6893,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.12186205215695832,
|
| 188 |
+
"grad_norm": 2.287115199507479,
|
| 189 |
+
"learning_rate": 9.985958293397433e-06,
|
| 190 |
+
"loss": 0.6801,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.12673653424323666,
|
| 195 |
+
"grad_norm": 2.111605184938528,
|
| 196 |
+
"learning_rate": 9.978855418031633e-06,
|
| 197 |
+
"loss": 0.6761,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.131611016329515,
|
| 202 |
+
"grad_norm": 2.3968266021989573,
|
| 203 |
+
"learning_rate": 9.970307445922905e-06,
|
| 204 |
+
"loss": 0.6659,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.13648549841579333,
|
| 209 |
+
"grad_norm": 2.0562856783183383,
|
| 210 |
+
"learning_rate": 9.960316858092613e-06,
|
| 211 |
+
"loss": 0.6735,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.14135998050207166,
|
| 216 |
+
"grad_norm": 2.2596310641814634,
|
| 217 |
+
"learning_rate": 9.948886554276689e-06,
|
| 218 |
+
"loss": 0.6658,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.14623446258835,
|
| 223 |
+
"grad_norm": 2.3851159107252022,
|
| 224 |
+
"learning_rate": 9.936019852083982e-06,
|
| 225 |
+
"loss": 0.6592,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.15110894467462832,
|
| 230 |
+
"grad_norm": 2.0161901435788083,
|
| 231 |
+
"learning_rate": 9.921720486033348e-06,
|
| 232 |
+
"loss": 0.657,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.15598342676090665,
|
| 237 |
+
"grad_norm": 2.127638747482369,
|
| 238 |
+
"learning_rate": 9.905992606469708e-06,
|
| 239 |
+
"loss": 0.6595,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.16085790884718498,
|
| 244 |
+
"grad_norm": 2.0531639641862447,
|
| 245 |
+
"learning_rate": 9.888840778359431e-06,
|
| 246 |
+
"loss": 0.6515,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1657323909334633,
|
| 251 |
+
"grad_norm": 1.8950315321987827,
|
| 252 |
+
"learning_rate": 9.870269979965364e-06,
|
| 253 |
+
"loss": 0.6492,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.17060687301974164,
|
| 258 |
+
"grad_norm": 1.9060347213321087,
|
| 259 |
+
"learning_rate": 9.850285601401899e-06,
|
| 260 |
+
"loss": 0.6458,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.17548135510601998,
|
| 265 |
+
"grad_norm": 2.074587641369688,
|
| 266 |
+
"learning_rate": 9.828893443070527e-06,
|
| 267 |
+
"loss": 0.6515,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1803558371922983,
|
| 272 |
+
"grad_norm": 2.021646896648869,
|
| 273 |
+
"learning_rate": 9.806099713976277e-06,
|
| 274 |
+
"loss": 0.6306,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.18523031927857664,
|
| 279 |
+
"grad_norm": 2.183178620435626,
|
| 280 |
+
"learning_rate": 9.781911029925573e-06,
|
| 281 |
+
"loss": 0.6317,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.190104801364855,
|
| 286 |
+
"grad_norm": 1.9176655344163256,
|
| 287 |
+
"learning_rate": 9.756334411606028e-06,
|
| 288 |
+
"loss": 0.6304,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.19497928345113333,
|
| 293 |
+
"grad_norm": 2.0283053056639226,
|
| 294 |
+
"learning_rate": 9.729377282548696e-06,
|
| 295 |
+
"loss": 0.6273,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.19985376553741166,
|
| 300 |
+
"grad_norm": 2.2614639215711794,
|
| 301 |
+
"learning_rate": 9.701047466973429e-06,
|
| 302 |
+
"loss": 0.6227,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.20472824762369,
|
| 307 |
+
"grad_norm": 2.048850431237601,
|
| 308 |
+
"learning_rate": 9.67135318751792e-06,
|
| 309 |
+
"loss": 0.6289,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.20960272970996832,
|
| 314 |
+
"grad_norm": 2.0716723168430224,
|
| 315 |
+
"learning_rate": 9.640303062851101e-06,
|
| 316 |
+
"loss": 0.6235,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.21447721179624665,
|
| 321 |
+
"grad_norm": 2.165745603614713,
|
| 322 |
+
"learning_rate": 9.607906105171613e-06,
|
| 323 |
+
"loss": 0.6254,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.21935169388252498,
|
| 328 |
+
"grad_norm": 2.121524753608251,
|
| 329 |
+
"learning_rate": 9.574171717592038e-06,
|
| 330 |
+
"loss": 0.6215,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.22422617596880332,
|
| 335 |
+
"grad_norm": 2.0155505697588207,
|
| 336 |
+
"learning_rate": 9.539109691409677e-06,
|
| 337 |
+
"loss": 0.6125,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.22910065805508165,
|
| 342 |
+
"grad_norm": 2.5497104730670457,
|
| 343 |
+
"learning_rate": 9.502730203264656e-06,
|
| 344 |
+
"loss": 0.6116,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.23397514014135998,
|
| 349 |
+
"grad_norm": 2.0859151782380163,
|
| 350 |
+
"learning_rate": 9.465043812186194e-06,
|
| 351 |
+
"loss": 0.6026,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2388496222276383,
|
| 356 |
+
"grad_norm": 2.0192086858289673,
|
| 357 |
+
"learning_rate": 9.426061456527871e-06,
|
| 358 |
+
"loss": 0.601,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.24372410431391664,
|
| 363 |
+
"grad_norm": 2.08123508728777,
|
| 364 |
+
"learning_rate": 9.385794450792818e-06,
|
| 365 |
+
"loss": 0.593,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.24859858640019497,
|
| 370 |
+
"grad_norm": 1.970692344326779,
|
| 371 |
+
"learning_rate": 9.344254482349702e-06,
|
| 372 |
+
"loss": 0.5879,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.25347306848647333,
|
| 377 |
+
"grad_norm": 2.0360574888538037,
|
| 378 |
+
"learning_rate": 9.301453608040523e-06,
|
| 379 |
+
"loss": 0.5884,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.25834755057275166,
|
| 384 |
+
"grad_norm": 1.9510711367676767,
|
| 385 |
+
"learning_rate": 9.25740425068114e-06,
|
| 386 |
+
"loss": 0.5937,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.26322203265903,
|
| 391 |
+
"grad_norm": 2.0679042358578057,
|
| 392 |
+
"learning_rate": 9.2121191954556e-06,
|
| 393 |
+
"loss": 0.5838,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2680965147453083,
|
| 398 |
+
"grad_norm": 2.0443212477081008,
|
| 399 |
+
"learning_rate": 9.165611586205268e-06,
|
| 400 |
+
"loss": 0.5719,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.27297099683158665,
|
| 405 |
+
"grad_norm": 2.0408583983165234,
|
| 406 |
+
"learning_rate": 9.11789492161388e-06,
|
| 407 |
+
"loss": 0.5839,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.277845478917865,
|
| 412 |
+
"grad_norm": 1.9767895517261123,
|
| 413 |
+
"learning_rate": 9.068983051289589e-06,
|
| 414 |
+
"loss": 0.5865,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2827199610041433,
|
| 419 |
+
"grad_norm": 1.881981157978116,
|
| 420 |
+
"learning_rate": 9.018890171745156e-06,
|
| 421 |
+
"loss": 0.5793,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.28759444309042165,
|
| 426 |
+
"grad_norm": 1.9932110390018152,
|
| 427 |
+
"learning_rate": 8.967630822277472e-06,
|
| 428 |
+
"loss": 0.5808,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2924689251767,
|
| 433 |
+
"grad_norm": 1.9915669044094475,
|
| 434 |
+
"learning_rate": 8.915219880747555e-06,
|
| 435 |
+
"loss": 0.569,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.2973434072629783,
|
| 440 |
+
"grad_norm": 2.046268209029942,
|
| 441 |
+
"learning_rate": 8.861672559262316e-06,
|
| 442 |
+
"loss": 0.5759,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.30221788934925664,
|
| 447 |
+
"grad_norm": 1.9535790327085676,
|
| 448 |
+
"learning_rate": 8.80700439975928e-06,
|
| 449 |
+
"loss": 0.5717,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.30709237143553497,
|
| 454 |
+
"grad_norm": 2.280776169449102,
|
| 455 |
+
"learning_rate": 8.751231269495604e-06,
|
| 456 |
+
"loss": 0.5703,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.3119668535218133,
|
| 461 |
+
"grad_norm": 2.0042263363709996,
|
| 462 |
+
"learning_rate": 8.694369356442638e-06,
|
| 463 |
+
"loss": 0.5668,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.31684133560809163,
|
| 468 |
+
"grad_norm": 2.098934601526615,
|
| 469 |
+
"learning_rate": 8.636435164587436e-06,
|
| 470 |
+
"loss": 0.5532,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.32171581769436997,
|
| 475 |
+
"grad_norm": 2.0675738967767496,
|
| 476 |
+
"learning_rate": 8.577445509142514e-06,
|
| 477 |
+
"loss": 0.5585,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3265902997806483,
|
| 482 |
+
"grad_norm": 2.3766255529170692,
|
| 483 |
+
"learning_rate": 8.517417511665299e-06,
|
| 484 |
+
"loss": 0.5658,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3314647818669266,
|
| 489 |
+
"grad_norm": 2.1910659775597248,
|
| 490 |
+
"learning_rate": 8.456368595088647e-06,
|
| 491 |
+
"loss": 0.5507,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.33633926395320496,
|
| 496 |
+
"grad_norm": 1.8703655986029497,
|
| 497 |
+
"learning_rate": 8.394316478663886e-06,
|
| 498 |
+
"loss": 0.5406,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.3412137460394833,
|
| 503 |
+
"grad_norm": 1.9244217565682906,
|
| 504 |
+
"learning_rate": 8.331279172817876e-06,
|
| 505 |
+
"loss": 0.542,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3460882281257616,
|
| 510 |
+
"grad_norm": 2.0014619556553335,
|
| 511 |
+
"learning_rate": 8.26727497392553e-06,
|
| 512 |
+
"loss": 0.5392,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.35096271021203995,
|
| 517 |
+
"grad_norm": 1.8442031091685414,
|
| 518 |
+
"learning_rate": 8.20232245899935e-06,
|
| 519 |
+
"loss": 0.5318,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3558371922983183,
|
| 524 |
+
"grad_norm": 1.9799423972998886,
|
| 525 |
+
"learning_rate": 8.136440480297514e-06,
|
| 526 |
+
"loss": 0.5414,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3607116743845966,
|
| 531 |
+
"grad_norm": 2.0577170932676596,
|
| 532 |
+
"learning_rate": 8.069648159852059e-06,
|
| 533 |
+
"loss": 0.5296,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.36558615647087495,
|
| 538 |
+
"grad_norm": 2.037446340676939,
|
| 539 |
+
"learning_rate": 8.001964883918793e-06,
|
| 540 |
+
"loss": 0.5348,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3704606385571533,
|
| 545 |
+
"grad_norm": 2.0638439360397665,
|
| 546 |
+
"learning_rate": 7.933410297350472e-06,
|
| 547 |
+
"loss": 0.5229,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3753351206434316,
|
| 552 |
+
"grad_norm": 1.9569671735850687,
|
| 553 |
+
"learning_rate": 7.864004297894963e-06,
|
| 554 |
+
"loss": 0.5275,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.38020960272971,
|
| 559 |
+
"grad_norm": 1.9870139157026745,
|
| 560 |
+
"learning_rate": 7.793767030419975e-06,
|
| 561 |
+
"loss": 0.533,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3850840848159883,
|
| 566 |
+
"grad_norm": 1.9631002389899173,
|
| 567 |
+
"learning_rate": 7.722718881066086e-06,
|
| 568 |
+
"loss": 0.5245,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.38995856690226666,
|
| 573 |
+
"grad_norm": 1.9193362092353006,
|
| 574 |
+
"learning_rate": 7.650880471329725e-06,
|
| 575 |
+
"loss": 0.5203,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.394833048988545,
|
| 580 |
+
"grad_norm": 1.9397872066961752,
|
| 581 |
+
"learning_rate": 7.578272652077849e-06,
|
| 582 |
+
"loss": 0.5144,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3997075310748233,
|
| 587 |
+
"grad_norm": 1.952394556800401,
|
| 588 |
+
"learning_rate": 7.504916497496051e-06,
|
| 589 |
+
"loss": 0.5168,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.40458201316110165,
|
| 594 |
+
"grad_norm": 1.9410715991990146,
|
| 595 |
+
"learning_rate": 7.43083329897184e-06,
|
| 596 |
+
"loss": 0.4964,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.40945649524738,
|
| 601 |
+
"grad_norm": 1.9371602423339187,
|
| 602 |
+
"learning_rate": 7.3560445589148875e-06,
|
| 603 |
+
"loss": 0.5136,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4143309773336583,
|
| 608 |
+
"grad_norm": 1.9398436533180357,
|
| 609 |
+
"learning_rate": 7.2805719845160195e-06,
|
| 610 |
+
"loss": 0.5012,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.41920545941993664,
|
| 615 |
+
"grad_norm": 1.8404891949351312,
|
| 616 |
+
"learning_rate": 7.20443748144678e-06,
|
| 617 |
+
"loss": 0.4985,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.424079941506215,
|
| 622 |
+
"grad_norm": 1.9219013215637166,
|
| 623 |
+
"learning_rate": 7.127663147501377e-06,
|
| 624 |
+
"loss": 0.497,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.4289544235924933,
|
| 629 |
+
"grad_norm": 2.087372537148497,
|
| 630 |
+
"learning_rate": 7.050271266182862e-06,
|
| 631 |
+
"loss": 0.4954,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.43382890567877164,
|
| 636 |
+
"grad_norm": 1.9732921364375713,
|
| 637 |
+
"learning_rate": 6.97228430023543e-06,
|
| 638 |
+
"loss": 0.4914,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.43870338776504997,
|
| 643 |
+
"grad_norm": 1.8509876619225545,
|
| 644 |
+
"learning_rate": 6.893724885124668e-06,
|
| 645 |
+
"loss": 0.4816,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4435778698513283,
|
| 650 |
+
"grad_norm": 1.876089892137146,
|
| 651 |
+
"learning_rate": 6.814615822467691e-06,
|
| 652 |
+
"loss": 0.4863,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.44845235193760663,
|
| 657 |
+
"grad_norm": 1.956376416630627,
|
| 658 |
+
"learning_rate": 6.734980073415038e-06,
|
| 659 |
+
"loss": 0.4914,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.45332683402388496,
|
| 664 |
+
"grad_norm": 1.831312024979652,
|
| 665 |
+
"learning_rate": 6.654840751986282e-06,
|
| 666 |
+
"loss": 0.4773,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.4582013161101633,
|
| 671 |
+
"grad_norm": 1.9745029935310336,
|
| 672 |
+
"learning_rate": 6.574221118361254e-06,
|
| 673 |
+
"loss": 0.4843,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.4630757981964416,
|
| 678 |
+
"grad_norm": 2.090848022257445,
|
| 679 |
+
"learning_rate": 6.493144572128852e-06,
|
| 680 |
+
"loss": 0.4891,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.46795028028271995,
|
| 685 |
+
"grad_norm": 2.0539193178381345,
|
| 686 |
+
"learning_rate": 6.411634645495388e-06,
|
| 687 |
+
"loss": 0.465,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4728247623689983,
|
| 692 |
+
"grad_norm": 1.8380774504527437,
|
| 693 |
+
"learning_rate": 6.329714996454436e-06,
|
| 694 |
+
"loss": 0.4717,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4776992444552766,
|
| 699 |
+
"grad_norm": 1.9501438562254991,
|
| 700 |
+
"learning_rate": 6.247409401920184e-06,
|
| 701 |
+
"loss": 0.47,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.48257372654155495,
|
| 706 |
+
"grad_norm": 1.9399214168055752,
|
| 707 |
+
"learning_rate": 6.164741750826246e-06,
|
| 708 |
+
"loss": 0.4696,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4874482086278333,
|
| 713 |
+
"grad_norm": 1.9438999474698049,
|
| 714 |
+
"learning_rate": 6.081736037191998e-06,
|
| 715 |
+
"loss": 0.4761,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.4923226907141116,
|
| 720 |
+
"grad_norm": 2.109991763270123,
|
| 721 |
+
"learning_rate": 5.998416353158369e-06,
|
| 722 |
+
"loss": 0.467,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.49719717280038994,
|
| 727 |
+
"grad_norm": 1.9305942803443825,
|
| 728 |
+
"learning_rate": 5.914806881995192e-06,
|
| 729 |
+
"loss": 0.4519,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5020716548866683,
|
| 734 |
+
"grad_norm": 1.8894659709352353,
|
| 735 |
+
"learning_rate": 5.830931891082077e-06,
|
| 736 |
+
"loss": 0.4625,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5069461369729467,
|
| 741 |
+
"grad_norm": 1.943600690573164,
|
| 742 |
+
"learning_rate": 5.746815724864884e-06,
|
| 743 |
+
"loss": 0.4486,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5118206190592249,
|
| 748 |
+
"grad_norm": 1.8604126506438579,
|
| 749 |
+
"learning_rate": 5.662482797789833e-06,
|
| 750 |
+
"loss": 0.4501,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5166951011455033,
|
| 755 |
+
"grad_norm": 1.8945174085661423,
|
| 756 |
+
"learning_rate": 5.577957587217281e-06,
|
| 757 |
+
"loss": 0.4576,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5215695832317816,
|
| 762 |
+
"grad_norm": 1.9454371455405457,
|
| 763 |
+
"learning_rate": 5.493264626317252e-06,
|
| 764 |
+
"loss": 0.4546,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.52644406531806,
|
| 769 |
+
"grad_norm": 1.9257763152448473,
|
| 770 |
+
"learning_rate": 5.408428496948761e-06,
|
| 771 |
+
"loss": 0.4476,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5313185474043383,
|
| 776 |
+
"grad_norm": 1.942601355689532,
|
| 777 |
+
"learning_rate": 5.323473822525011e-06,
|
| 778 |
+
"loss": 0.4419,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5361930294906166,
|
| 783 |
+
"grad_norm": 1.8380797740924733,
|
| 784 |
+
"learning_rate": 5.238425260866524e-06,
|
| 785 |
+
"loss": 0.4339,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5410675115768949,
|
| 790 |
+
"grad_norm": 1.9856061522664756,
|
| 791 |
+
"learning_rate": 5.153307497044291e-06,
|
| 792 |
+
"loss": 0.4486,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5459419936631733,
|
| 797 |
+
"grad_norm": 2.0152615373656446,
|
| 798 |
+
"learning_rate": 5.068145236215007e-06,
|
| 799 |
+
"loss": 0.4361,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5508164757494516,
|
| 804 |
+
"grad_norm": 1.8713611793445957,
|
| 805 |
+
"learning_rate": 4.982963196450478e-06,
|
| 806 |
+
"loss": 0.4388,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.55569095783573,
|
| 811 |
+
"grad_norm": 1.9591094110255243,
|
| 812 |
+
"learning_rate": 4.8977861015632865e-06,
|
| 813 |
+
"loss": 0.4382,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5605654399220082,
|
| 818 |
+
"grad_norm": 1.8237556125663812,
|
| 819 |
+
"learning_rate": 4.812638673930777e-06,
|
| 820 |
+
"loss": 0.4289,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5654399220082866,
|
| 825 |
+
"grad_norm": 1.8090312867754712,
|
| 826 |
+
"learning_rate": 4.72754562731947e-06,
|
| 827 |
+
"loss": 0.4258,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5703144040945649,
|
| 832 |
+
"grad_norm": 1.8257902996618338,
|
| 833 |
+
"learning_rate": 4.64253165971197e-06,
|
| 834 |
+
"loss": 0.4303,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5751888861808433,
|
| 839 |
+
"grad_norm": 2.0254032476042223,
|
| 840 |
+
"learning_rate": 4.557621446138455e-06,
|
| 841 |
+
"loss": 0.4202,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5800633682671216,
|
| 846 |
+
"grad_norm": 1.9947879579802112,
|
| 847 |
+
"learning_rate": 4.47283963151483e-06,
|
| 848 |
+
"loss": 0.424,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5849378503534,
|
| 853 |
+
"grad_norm": 1.8982751664240438,
|
| 854 |
+
"learning_rate": 4.388210823489616e-06,
|
| 855 |
+
"loss": 0.4221,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5898123324396782,
|
| 860 |
+
"grad_norm": 1.8999729869459319,
|
| 861 |
+
"learning_rate": 4.3037595853016645e-06,
|
| 862 |
+
"loss": 0.4162,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5946868145259566,
|
| 867 |
+
"grad_norm": 1.9051454199437885,
|
| 868 |
+
"learning_rate": 4.219510428650752e-06,
|
| 869 |
+
"loss": 0.4154,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5995612966122349,
|
| 874 |
+
"grad_norm": 1.8892938847030196,
|
| 875 |
+
"learning_rate": 4.135487806583141e-06,
|
| 876 |
+
"loss": 0.4183,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6044357786985133,
|
| 881 |
+
"grad_norm": 1.8391212463048745,
|
| 882 |
+
"learning_rate": 4.051716106394162e-06,
|
| 883 |
+
"loss": 0.4169,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6093102607847917,
|
| 888 |
+
"grad_norm": 1.9164931200596216,
|
| 889 |
+
"learning_rate": 3.968219642549876e-06,
|
| 890 |
+
"loss": 0.4096,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.6141847428710699,
|
| 895 |
+
"grad_norm": 1.8882587070789514,
|
| 896 |
+
"learning_rate": 3.885022649629887e-06,
|
| 897 |
+
"loss": 0.4089,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6190592249573483,
|
| 902 |
+
"grad_norm": 2.0414351606801677,
|
| 903 |
+
"learning_rate": 3.8021492752933196e-06,
|
| 904 |
+
"loss": 0.4118,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6239337070436266,
|
| 909 |
+
"grad_norm": 1.88948983848305,
|
| 910 |
+
"learning_rate": 3.7196235732700546e-06,
|
| 911 |
+
"loss": 0.4155,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.628808189129905,
|
| 916 |
+
"grad_norm": 1.9185526120999834,
|
| 917 |
+
"learning_rate": 3.637469496379201e-06,
|
| 918 |
+
"loss": 0.3988,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6336826712161833,
|
| 923 |
+
"grad_norm": 1.8441572398843216,
|
| 924 |
+
"learning_rate": 3.5557108895768723e-06,
|
| 925 |
+
"loss": 0.4099,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6385571533024617,
|
| 930 |
+
"grad_norm": 2.0485958888309796,
|
| 931 |
+
"learning_rate": 3.4743714830352604e-06,
|
| 932 |
+
"loss": 0.405,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6434316353887399,
|
| 937 |
+
"grad_norm": 1.806992701143073,
|
| 938 |
+
"learning_rate": 3.3934748852550353e-06,
|
| 939 |
+
"loss": 0.4037,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.6483061174750183,
|
| 944 |
+
"grad_norm": 1.943526238424205,
|
| 945 |
+
"learning_rate": 3.3130445762130485e-06,
|
| 946 |
+
"loss": 0.3967,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.6531805995612966,
|
| 951 |
+
"grad_norm": 1.8848557764463123,
|
| 952 |
+
"learning_rate": 3.2331039005473495e-06,
|
| 953 |
+
"loss": 0.3924,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.658055081647575,
|
| 958 |
+
"grad_norm": 1.892841917353598,
|
| 959 |
+
"learning_rate": 3.1536760607814747e-06,
|
| 960 |
+
"loss": 0.3961,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6629295637338533,
|
| 965 |
+
"grad_norm": 1.7833910365050067,
|
| 966 |
+
"learning_rate": 3.0747841105899965e-06,
|
| 967 |
+
"loss": 0.3973,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.6678040458201316,
|
| 972 |
+
"grad_norm": 1.8266463006432494,
|
| 973 |
+
"learning_rate": 2.9964509481072627e-06,
|
| 974 |
+
"loss": 0.3829,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.6726785279064099,
|
| 979 |
+
"grad_norm": 1.9794830429310166,
|
| 980 |
+
"learning_rate": 2.918699309281292e-06,
|
| 981 |
+
"loss": 0.3886,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.6775530099926883,
|
| 986 |
+
"grad_norm": 1.8605018457416864,
|
| 987 |
+
"learning_rate": 2.84155176127473e-06,
|
| 988 |
+
"loss": 0.3889,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6824274920789666,
|
| 993 |
+
"grad_norm": 1.8677944161418016,
|
| 994 |
+
"learning_rate": 2.765030695914815e-06,
|
| 995 |
+
"loss": 0.3878,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.687301974165245,
|
| 1000 |
+
"grad_norm": 1.7809933219767806,
|
| 1001 |
+
"learning_rate": 2.689158323194212e-06,
|
| 1002 |
+
"loss": 0.389,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6921764562515232,
|
| 1007 |
+
"grad_norm": 1.9534924091406372,
|
| 1008 |
+
"learning_rate": 2.6139566648246355e-06,
|
| 1009 |
+
"loss": 0.38,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6970509383378016,
|
| 1014 |
+
"grad_norm": 1.8154536663255623,
|
| 1015 |
+
"learning_rate": 2.5394475478451246e-06,
|
| 1016 |
+
"loss": 0.3819,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7019254204240799,
|
| 1021 |
+
"grad_norm": 1.910488831165307,
|
| 1022 |
+
"learning_rate": 2.4656525982868106e-06,
|
| 1023 |
+
"loss": 0.3805,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7067999025103583,
|
| 1028 |
+
"grad_norm": 1.8313054725277078,
|
| 1029 |
+
"learning_rate": 2.39259323489603e-06,
|
| 1030 |
+
"loss": 0.3742,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.7116743845966366,
|
| 1035 |
+
"grad_norm": 1.8851239942706999,
|
| 1036 |
+
"learning_rate": 2.320290662917607e-06,
|
| 1037 |
+
"loss": 0.3726,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.716548866682915,
|
| 1042 |
+
"grad_norm": 1.8525938618294533,
|
| 1043 |
+
"learning_rate": 2.2487658679400943e-06,
|
| 1044 |
+
"loss": 0.3812,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7214233487691932,
|
| 1049 |
+
"grad_norm": 1.952969026800498,
|
| 1050 |
+
"learning_rate": 2.178039609804777e-06,
|
| 1051 |
+
"loss": 0.3757,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7262978308554716,
|
| 1056 |
+
"grad_norm": 1.8641758450652162,
|
| 1057 |
+
"learning_rate": 2.108132416580198e-06,
|
| 1058 |
+
"loss": 0.3794,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7311723129417499,
|
| 1063 |
+
"grad_norm": 1.8518338990761232,
|
| 1064 |
+
"learning_rate": 2.0390645786039406e-06,
|
| 1065 |
+
"loss": 0.3713,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7360467950280283,
|
| 1070 |
+
"grad_norm": 1.8015066231257286,
|
| 1071 |
+
"learning_rate": 1.9708561425934393e-06,
|
| 1072 |
+
"loss": 0.3784,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7409212771143066,
|
| 1077 |
+
"grad_norm": 1.8409794783180748,
|
| 1078 |
+
"learning_rate": 1.903526905827474e-06,
|
| 1079 |
+
"loss": 0.3751,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.7457957592005849,
|
| 1084 |
+
"grad_norm": 1.8641076874515972,
|
| 1085 |
+
"learning_rate": 1.8370964104000783e-06,
|
| 1086 |
+
"loss": 0.3746,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.7506702412868632,
|
| 1091 |
+
"grad_norm": 1.786737876865444,
|
| 1092 |
+
"learning_rate": 1.7715839375485067e-06,
|
| 1093 |
+
"loss": 0.3628,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.7555447233731416,
|
| 1098 |
+
"grad_norm": 1.8446799495571504,
|
| 1099 |
+
"learning_rate": 1.7070085020569194e-06,
|
| 1100 |
+
"loss": 0.3644,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.76041920545942,
|
| 1105 |
+
"grad_norm": 1.751658260263151,
|
| 1106 |
+
"learning_rate": 1.6433888467374015e-06,
|
| 1107 |
+
"loss": 0.37,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.7652936875456983,
|
| 1112 |
+
"grad_norm": 1.8477843728803192,
|
| 1113 |
+
"learning_rate": 1.5807434369899248e-06,
|
| 1114 |
+
"loss": 0.3628,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.7701681696319767,
|
| 1119 |
+
"grad_norm": 1.8593690412561963,
|
| 1120 |
+
"learning_rate": 1.51909045544282e-06,
|
| 1121 |
+
"loss": 0.3708,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.7750426517182549,
|
| 1126 |
+
"grad_norm": 1.747782138054366,
|
| 1127 |
+
"learning_rate": 1.4584477966753324e-06,
|
| 1128 |
+
"loss": 0.3652,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.7799171338045333,
|
| 1133 |
+
"grad_norm": 1.755667670884649,
|
| 1134 |
+
"learning_rate": 1.398833062023775e-06,
|
| 1135 |
+
"loss": 0.3691,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.7847916158908116,
|
| 1140 |
+
"grad_norm": 1.8425636184668681,
|
| 1141 |
+
"learning_rate": 1.3402635544727992e-06,
|
| 1142 |
+
"loss": 0.366,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.78966609797709,
|
| 1147 |
+
"grad_norm": 1.9026395556203364,
|
| 1148 |
+
"learning_rate": 1.2827562736332555e-06,
|
| 1149 |
+
"loss": 0.3589,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.7945405800633683,
|
| 1154 |
+
"grad_norm": 1.7133796113318556,
|
| 1155 |
+
"learning_rate": 1.226327910808116e-06,
|
| 1156 |
+
"loss": 0.3597,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.7994150621496466,
|
| 1161 |
+
"grad_norm": 1.8616850952101744,
|
| 1162 |
+
"learning_rate": 1.1709948441478763e-06,
|
| 1163 |
+
"loss": 0.3583,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8042895442359249,
|
| 1168 |
+
"grad_norm": 1.7675149524570417,
|
| 1169 |
+
"learning_rate": 1.116773133896848e-06,
|
| 1170 |
+
"loss": 0.3605,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8091640263222033,
|
| 1175 |
+
"grad_norm": 1.900735872689697,
|
| 1176 |
+
"learning_rate": 1.0636785177317255e-06,
|
| 1177 |
+
"loss": 0.3547,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8140385084084816,
|
| 1182 |
+
"grad_norm": 1.7353000724393655,
|
| 1183 |
+
"learning_rate": 1.0117264061937777e-06,
|
| 1184 |
+
"loss": 0.3543,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.81891299049476,
|
| 1189 |
+
"grad_norm": 1.8226701532638399,
|
| 1190 |
+
"learning_rate": 9.60931878215985e-07,
|
| 1191 |
+
"loss": 0.3523,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.8237874725810382,
|
| 1196 |
+
"grad_norm": 1.7199982449094653,
|
| 1197 |
+
"learning_rate": 9.113096767464302e-07,
|
| 1198 |
+
"loss": 0.3572,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.8286619546673166,
|
| 1203 |
+
"grad_norm": 1.7416045554651556,
|
| 1204 |
+
"learning_rate": 8.62874204469204e-07,
|
| 1205 |
+
"loss": 0.3546,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.8335364367535949,
|
| 1210 |
+
"grad_norm": 1.7748467098472374,
|
| 1211 |
+
"learning_rate": 8.156395196240752e-07,
|
| 1212 |
+
"loss": 0.3488,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8384109188398733,
|
| 1217 |
+
"grad_norm": 1.8256676668006175,
|
| 1218 |
+
"learning_rate": 7.696193319261242e-07,
|
| 1219 |
+
"loss": 0.3467,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.8432854009261516,
|
| 1224 |
+
"grad_norm": 1.8827284517179357,
|
| 1225 |
+
"learning_rate": 7.248269985865514e-07,
|
| 1226 |
+
"loss": 0.3521,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.84815988301243,
|
| 1231 |
+
"grad_norm": 1.6781224427131647,
|
| 1232 |
+
"learning_rate": 6.812755204357857e-07,
|
| 1233 |
+
"loss": 0.3535,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.8530343650987082,
|
| 1238 |
+
"grad_norm": 1.6995357059191782,
|
| 1239 |
+
"learning_rate": 6.389775381500351e-07,
|
| 1240 |
+
"loss": 0.3435,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.8579088471849866,
|
| 1245 |
+
"grad_norm": 1.7402333449103136,
|
| 1246 |
+
"learning_rate": 5.979453285823711e-07,
|
| 1247 |
+
"loss": 0.3443,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.8627833292712649,
|
| 1252 |
+
"grad_norm": 1.7615298544617708,
|
| 1253 |
+
"learning_rate": 5.58190801199413e-07,
|
| 1254 |
+
"loss": 0.3528,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.8676578113575433,
|
| 1259 |
+
"grad_norm": 1.7343491942185678,
|
| 1260 |
+
"learning_rate": 5.197254946246416e-07,
|
| 1261 |
+
"loss": 0.3495,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.8725322934438215,
|
| 1266 |
+
"grad_norm": 1.657555480865734,
|
| 1267 |
+
"learning_rate": 4.825605732893546e-07,
|
| 1268 |
+
"loss": 0.3468,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.8774067755300999,
|
| 1273 |
+
"grad_norm": 1.6499922538828395,
|
| 1274 |
+
"learning_rate": 4.4670682419221954e-07,
|
| 1275 |
+
"loss": 0.3396,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.8822812576163782,
|
| 1280 |
+
"grad_norm": 1.7518714860746227,
|
| 1281 |
+
"learning_rate": 4.121746537683907e-07,
|
| 1282 |
+
"loss": 0.3504,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.8871557397026566,
|
| 1287 |
+
"grad_norm": 1.770832100980816,
|
| 1288 |
+
"learning_rate": 3.789740848690682e-07,
|
| 1289 |
+
"loss": 0.3518,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.8920302217889349,
|
| 1294 |
+
"grad_norm": 1.7567597669594073,
|
| 1295 |
+
"learning_rate": 3.4711475385240057e-07,
|
| 1296 |
+
"loss": 0.3492,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.8969047038752133,
|
| 1301 |
+
"grad_norm": 1.7161657861744943,
|
| 1302 |
+
"learning_rate": 3.1660590778656406e-07,
|
| 1303 |
+
"loss": 0.3428,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9017791859614915,
|
| 1308 |
+
"grad_norm": 1.728233513619595,
|
| 1309 |
+
"learning_rate": 2.8745640176582766e-07,
|
| 1310 |
+
"loss": 0.3396,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.9066536680477699,
|
| 1315 |
+
"grad_norm": 1.8381612957565303,
|
| 1316 |
+
"learning_rate": 2.5967469634039177e-07,
|
| 1317 |
+
"loss": 0.345,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9115281501340483,
|
| 1322 |
+
"grad_norm": 1.7243006426458327,
|
| 1323 |
+
"learning_rate": 2.3326885506074314e-07,
|
| 1324 |
+
"loss": 0.3465,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.9164026322203266,
|
| 1329 |
+
"grad_norm": 1.5970738101828363,
|
| 1330 |
+
"learning_rate": 2.0824654213723038e-07,
|
| 1331 |
+
"loss": 0.3429,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.921277114306605,
|
| 1336 |
+
"grad_norm": 1.6533959425871336,
|
| 1337 |
+
"learning_rate": 1.8461502021555721e-07,
|
| 1338 |
+
"loss": 0.3389,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.9261515963928832,
|
| 1343 |
+
"grad_norm": 1.7738912948147607,
|
| 1344 |
+
"learning_rate": 1.6238114826881868e-07,
|
| 1345 |
+
"loss": 0.3439,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.9310260784791616,
|
| 1350 |
+
"grad_norm": 1.7429674496591296,
|
| 1351 |
+
"learning_rate": 1.4155137960670974e-07,
|
| 1352 |
+
"loss": 0.3389,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.9359005605654399,
|
| 1357 |
+
"grad_norm": 1.7343774559673155,
|
| 1358 |
+
"learning_rate": 1.2213176000246852e-07,
|
| 1359 |
+
"loss": 0.345,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.9407750426517183,
|
| 1364 |
+
"grad_norm": 1.7876244818652967,
|
| 1365 |
+
"learning_rate": 1.0412792593811505e-07,
|
| 1366 |
+
"loss": 0.3426,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.9456495247379966,
|
| 1371 |
+
"grad_norm": 1.6810981829120015,
|
| 1372 |
+
"learning_rate": 8.754510296847651e-08,
|
| 1373 |
+
"loss": 0.3444,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.950524006824275,
|
| 1378 |
+
"grad_norm": 1.7314637174387817,
|
| 1379 |
+
"learning_rate": 7.238810420448883e-08,
|
| 1380 |
+
"loss": 0.3339,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.9553984889105532,
|
| 1385 |
+
"grad_norm": 1.6057497995450185,
|
| 1386 |
+
"learning_rate": 5.866132891620746e-08,
|
| 1387 |
+
"loss": 0.346,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.9602729709968316,
|
| 1392 |
+
"grad_norm": 1.7119707817106466,
|
| 1393 |
+
"learning_rate": 4.6368761255930485e-08,
|
| 1394 |
+
"loss": 0.3379,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.9651474530831099,
|
| 1399 |
+
"grad_norm": 1.7236174016491694,
|
| 1400 |
+
"learning_rate": 3.551396910181415e-08,
|
| 1401 |
+
"loss": 0.3412,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.9700219351693883,
|
| 1406 |
+
"grad_norm": 1.6591399917697027,
|
| 1407 |
+
"learning_rate": 2.6100103022306257e-08,
|
| 1408 |
+
"loss": 0.3433,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.9748964172556666,
|
| 1413 |
+
"grad_norm": 1.7276142814853213,
|
| 1414 |
+
"learning_rate": 1.812989536170484e-08,
|
| 1415 |
+
"loss": 0.3414,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.979770899341945,
|
| 1420 |
+
"grad_norm": 1.6558936839338774,
|
| 1421 |
+
"learning_rate": 1.1605659447102568e-08,
|
| 1422 |
+
"loss": 0.3366,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.9846453814282232,
|
| 1427 |
+
"grad_norm": 1.6766541339225858,
|
| 1428 |
+
"learning_rate": 6.529288916952703e-09,
|
| 1429 |
+
"loss": 0.3331,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.9895198635145016,
|
| 1434 |
+
"grad_norm": 1.7452767005894518,
|
| 1435 |
+
"learning_rate": 2.9022571714448776e-09,
|
| 1436 |
+
"loss": 0.3429,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.9943943456007799,
|
| 1441 |
+
"grad_norm": 1.8097610241635727,
|
| 1442 |
+
"learning_rate": 7.256169448560668e-10,
|
| 1443 |
+
"loss": 0.3415,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.9992688276870583,
|
| 1448 |
+
"grad_norm": 1.758337391190151,
|
| 1449 |
+
"learning_rate": 0.0,
|
| 1450 |
+
"loss": 0.3467,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.9992688276870583,
|
| 1455 |
+
"eval_loss": 0.3359443247318268,
|
| 1456 |
+
"eval_runtime": 96.9406,
|
| 1457 |
+
"eval_samples_per_second": 3.115,
|
| 1458 |
+
"eval_steps_per_second": 0.784,
|
| 1459 |
+
"step": 1025
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.9992688276870583,
|
| 1463 |
+
"step": 1025,
|
| 1464 |
+
"total_flos": 214561802158080.0,
|
| 1465 |
+
"train_loss": 0.5001517156275307,
|
| 1466 |
+
"train_runtime": 26987.6611,
|
| 1467 |
+
"train_samples_per_second": 1.216,
|
| 1468 |
+
"train_steps_per_second": 0.038
|
| 1469 |
+
}
|
| 1470 |
+
],
|
| 1471 |
+
"logging_steps": 5,
|
| 1472 |
+
"max_steps": 1025,
|
| 1473 |
+
"num_input_tokens_seen": 0,
|
| 1474 |
+
"num_train_epochs": 1,
|
| 1475 |
+
"save_steps": 100,
|
| 1476 |
+
"stateful_callbacks": {
|
| 1477 |
+
"TrainerControl": {
|
| 1478 |
+
"args": {
|
| 1479 |
+
"should_epoch_stop": false,
|
| 1480 |
+
"should_evaluate": false,
|
| 1481 |
+
"should_log": false,
|
| 1482 |
+
"should_save": true,
|
| 1483 |
+
"should_training_stop": true
|
| 1484 |
+
},
|
| 1485 |
+
"attributes": {}
|
| 1486 |
+
}
|
| 1487 |
+
},
|
| 1488 |
+
"total_flos": 214561802158080.0,
|
| 1489 |
+
"train_batch_size": 2,
|
| 1490 |
+
"trial_name": null,
|
| 1491 |
+
"trial_params": null
|
| 1492 |
+
}
|