Model save
Browse files- README.md +69 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1513 -0
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama3
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: downstream-7b-p0.1_seed42_rare
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# downstream-7b-p0.1_seed42_rare
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.0771
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0002
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 1
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 4
|
| 48 |
+
- gradient_accumulation_steps: 4
|
| 49 |
+
- total_train_batch_size: 64
|
| 50 |
+
- total_eval_batch_size: 4
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 1
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 1.0444 | 0.9993 | 1044 | 1.0771 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.11.1
|
| 66 |
+
- Transformers 4.43.4
|
| 67 |
+
- Pytorch 2.3.1+cu121
|
| 68 |
+
- Datasets 2.19.1
|
| 69 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9992821249102656,
|
| 3 |
+
"total_flos": 2155604625850368.0,
|
| 4 |
+
"train_loss": 1.091635976486279,
|
| 5 |
+
"train_runtime": 24351.9745,
|
| 6 |
+
"train_samples": 114325,
|
| 7 |
+
"train_samples_per_second": 2.746,
|
| 8 |
+
"train_steps_per_second": 0.043
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9992821249102656,
|
| 3 |
+
"total_flos": 2155604625850368.0,
|
| 4 |
+
"train_loss": 1.091635976486279,
|
| 5 |
+
"train_runtime": 24351.9745,
|
| 6 |
+
"train_samples": 114325,
|
| 7 |
+
"train_samples_per_second": 2.746,
|
| 8 |
+
"train_steps_per_second": 0.043
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1513 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9992821249102656,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1044,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000957166786312515,
|
| 13 |
+
"grad_norm": 3.9142152723532337,
|
| 14 |
+
"learning_rate": 1.9047619047619051e-06,
|
| 15 |
+
"loss": 1.3978,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004785833931562575,
|
| 20 |
+
"grad_norm": 1.3430217420733046,
|
| 21 |
+
"learning_rate": 9.523809523809523e-06,
|
| 22 |
+
"loss": 1.3489,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.00957166786312515,
|
| 27 |
+
"grad_norm": 0.5850408636793494,
|
| 28 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 29 |
+
"loss": 1.2871,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014357501794687724,
|
| 34 |
+
"grad_norm": 0.46666716038967326,
|
| 35 |
+
"learning_rate": 2.857142857142857e-05,
|
| 36 |
+
"loss": 1.2106,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0191433357262503,
|
| 41 |
+
"grad_norm": 0.35044064248530404,
|
| 42 |
+
"learning_rate": 3.809523809523809e-05,
|
| 43 |
+
"loss": 1.189,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.023929169657812874,
|
| 48 |
+
"grad_norm": 0.27361957875198517,
|
| 49 |
+
"learning_rate": 4.761904761904762e-05,
|
| 50 |
+
"loss": 1.1469,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.028715003589375447,
|
| 55 |
+
"grad_norm": 0.2368453005937937,
|
| 56 |
+
"learning_rate": 5.714285714285714e-05,
|
| 57 |
+
"loss": 1.158,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03350083752093802,
|
| 62 |
+
"grad_norm": 0.2277332385016794,
|
| 63 |
+
"learning_rate": 6.666666666666667e-05,
|
| 64 |
+
"loss": 1.1437,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0382866714525006,
|
| 69 |
+
"grad_norm": 0.2265653549311157,
|
| 70 |
+
"learning_rate": 7.619047619047618e-05,
|
| 71 |
+
"loss": 1.1302,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.043072505384063174,
|
| 76 |
+
"grad_norm": 0.22079711284915807,
|
| 77 |
+
"learning_rate": 8.571428571428571e-05,
|
| 78 |
+
"loss": 1.13,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04785833931562575,
|
| 83 |
+
"grad_norm": 0.20813516832540208,
|
| 84 |
+
"learning_rate": 9.523809523809524e-05,
|
| 85 |
+
"loss": 1.106,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05264417324718832,
|
| 90 |
+
"grad_norm": 0.2044131638757028,
|
| 91 |
+
"learning_rate": 0.00010476190476190477,
|
| 92 |
+
"loss": 1.1348,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.057430007178750894,
|
| 97 |
+
"grad_norm": 0.20101729146508107,
|
| 98 |
+
"learning_rate": 0.00011428571428571428,
|
| 99 |
+
"loss": 1.1018,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.062215841110313475,
|
| 104 |
+
"grad_norm": 0.21865369935553125,
|
| 105 |
+
"learning_rate": 0.0001238095238095238,
|
| 106 |
+
"loss": 1.1129,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06700167504187604,
|
| 111 |
+
"grad_norm": 0.18405578482864565,
|
| 112 |
+
"learning_rate": 0.00013333333333333334,
|
| 113 |
+
"loss": 1.1018,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.07178750897343862,
|
| 118 |
+
"grad_norm": 0.18488079729650672,
|
| 119 |
+
"learning_rate": 0.00014285714285714287,
|
| 120 |
+
"loss": 1.1417,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0765733429050012,
|
| 125 |
+
"grad_norm": 0.18433481759594844,
|
| 126 |
+
"learning_rate": 0.00015238095238095237,
|
| 127 |
+
"loss": 1.111,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.08135917683656377,
|
| 132 |
+
"grad_norm": 0.20377971597879482,
|
| 133 |
+
"learning_rate": 0.00016190476190476192,
|
| 134 |
+
"loss": 1.0709,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08614501076812635,
|
| 139 |
+
"grad_norm": 0.20225554382239913,
|
| 140 |
+
"learning_rate": 0.00017142857142857143,
|
| 141 |
+
"loss": 1.1142,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.09093084469968891,
|
| 146 |
+
"grad_norm": 0.18520967333311886,
|
| 147 |
+
"learning_rate": 0.00018095238095238095,
|
| 148 |
+
"loss": 1.1142,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0957166786312515,
|
| 153 |
+
"grad_norm": 0.19606367225373053,
|
| 154 |
+
"learning_rate": 0.00019047619047619048,
|
| 155 |
+
"loss": 1.1049,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.10050251256281408,
|
| 160 |
+
"grad_norm": 0.1867473714189168,
|
| 161 |
+
"learning_rate": 0.0002,
|
| 162 |
+
"loss": 1.0927,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.10528834649437664,
|
| 167 |
+
"grad_norm": 0.185817071062854,
|
| 168 |
+
"learning_rate": 0.00019998600836567816,
|
| 169 |
+
"loss": 1.1206,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.11007418042593922,
|
| 174 |
+
"grad_norm": 0.1762396939142846,
|
| 175 |
+
"learning_rate": 0.00019994403737802927,
|
| 176 |
+
"loss": 1.1022,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.11486001435750179,
|
| 181 |
+
"grad_norm": 0.16668242539032083,
|
| 182 |
+
"learning_rate": 0.00019987409878190752,
|
| 183 |
+
"loss": 1.1052,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.11964584828906437,
|
| 188 |
+
"grad_norm": 0.1764182848939352,
|
| 189 |
+
"learning_rate": 0.00019977621214841822,
|
| 190 |
+
"loss": 1.1059,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.12443168222062695,
|
| 195 |
+
"grad_norm": 0.1787380695504439,
|
| 196 |
+
"learning_rate": 0.0001996504048694409,
|
| 197 |
+
"loss": 1.1102,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.12921751615218952,
|
| 202 |
+
"grad_norm": 0.18152638872711746,
|
| 203 |
+
"learning_rate": 0.00019949671214996445,
|
| 204 |
+
"loss": 1.0986,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.13400335008375208,
|
| 209 |
+
"grad_norm": 0.1745817045352934,
|
| 210 |
+
"learning_rate": 0.00019931517699823547,
|
| 211 |
+
"loss": 1.085,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.13878918401531468,
|
| 216 |
+
"grad_norm": 0.17554030590687575,
|
| 217 |
+
"learning_rate": 0.0001991058502137231,
|
| 218 |
+
"loss": 1.1363,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.14357501794687724,
|
| 223 |
+
"grad_norm": 0.1800405053641118,
|
| 224 |
+
"learning_rate": 0.00019886879037290384,
|
| 225 |
+
"loss": 1.0924,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.1483608518784398,
|
| 230 |
+
"grad_norm": 0.19148632700424054,
|
| 231 |
+
"learning_rate": 0.0001986040638128698,
|
| 232 |
+
"loss": 1.0824,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.1531466858100024,
|
| 237 |
+
"grad_norm": 0.17313730878343153,
|
| 238 |
+
"learning_rate": 0.0001983117446127654,
|
| 239 |
+
"loss": 1.1071,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.15793251974156497,
|
| 244 |
+
"grad_norm": 0.16961371492797625,
|
| 245 |
+
"learning_rate": 0.00019799191457305768,
|
| 246 |
+
"loss": 1.1311,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.16271835367312754,
|
| 251 |
+
"grad_norm": 0.17327733992734365,
|
| 252 |
+
"learning_rate": 0.00019764466319264595,
|
| 253 |
+
"loss": 1.1133,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.16750418760469013,
|
| 258 |
+
"grad_norm": 0.17176717180251114,
|
| 259 |
+
"learning_rate": 0.00019727008764381675,
|
| 260 |
+
"loss": 1.1153,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1722900215362527,
|
| 265 |
+
"grad_norm": 0.17138393560502058,
|
| 266 |
+
"learning_rate": 0.0001968682927450523,
|
| 267 |
+
"loss": 1.1006,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.17707585546781526,
|
| 272 |
+
"grad_norm": 0.16465106542109514,
|
| 273 |
+
"learning_rate": 0.00019643939093169844,
|
| 274 |
+
"loss": 1.104,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.18186168939937783,
|
| 279 |
+
"grad_norm": 0.1699157446206454,
|
| 280 |
+
"learning_rate": 0.00019598350222450178,
|
| 281 |
+
"loss": 1.1167,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.18664752333094042,
|
| 286 |
+
"grad_norm": 0.17602023031266878,
|
| 287 |
+
"learning_rate": 0.00019550075419602408,
|
| 288 |
+
"loss": 1.1131,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.191433357262503,
|
| 293 |
+
"grad_norm": 0.18282225583073394,
|
| 294 |
+
"learning_rate": 0.00019499128193494297,
|
| 295 |
+
"loss": 1.0889,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.19621919119406556,
|
| 300 |
+
"grad_norm": 0.16528930064048408,
|
| 301 |
+
"learning_rate": 0.0001944552280082499,
|
| 302 |
+
"loss": 1.1013,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.20100502512562815,
|
| 307 |
+
"grad_norm": 0.16504045631379008,
|
| 308 |
+
"learning_rate": 0.0001938927424213553,
|
| 309 |
+
"loss": 1.1003,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.20579085905719072,
|
| 314 |
+
"grad_norm": 0.16559761424705557,
|
| 315 |
+
"learning_rate": 0.000193303982576112,
|
| 316 |
+
"loss": 1.0998,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.21057669298875328,
|
| 321 |
+
"grad_norm": 0.16791680669282769,
|
| 322 |
+
"learning_rate": 0.0001926891132267692,
|
| 323 |
+
"loss": 1.0919,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.21536252692031588,
|
| 328 |
+
"grad_norm": 0.16689008275055853,
|
| 329 |
+
"learning_rate": 0.00019204830643386868,
|
| 330 |
+
"loss": 1.1069,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.22014836085187844,
|
| 335 |
+
"grad_norm": 0.17305744880787605,
|
| 336 |
+
"learning_rate": 0.00019138174151609683,
|
| 337 |
+
"loss": 1.1272,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.224934194783441,
|
| 342 |
+
"grad_norm": 0.1618964004882894,
|
| 343 |
+
"learning_rate": 0.00019068960500010523,
|
| 344 |
+
"loss": 1.0827,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.22972002871500358,
|
| 349 |
+
"grad_norm": 0.16411871929076272,
|
| 350 |
+
"learning_rate": 0.00018997209056831462,
|
| 351 |
+
"loss": 1.1164,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.23450586264656617,
|
| 356 |
+
"grad_norm": 0.1691454723246668,
|
| 357 |
+
"learning_rate": 0.0001892293990047159,
|
| 358 |
+
"loss": 1.1079,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.23929169657812874,
|
| 363 |
+
"grad_norm": 0.1649265891086064,
|
| 364 |
+
"learning_rate": 0.00018846173813868454,
|
| 365 |
+
"loss": 1.0825,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2440775305096913,
|
| 370 |
+
"grad_norm": 0.17018378559137046,
|
| 371 |
+
"learning_rate": 0.000187669322786823,
|
| 372 |
+
"loss": 1.1216,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2488633644412539,
|
| 377 |
+
"grad_norm": 0.1703316481606123,
|
| 378 |
+
"learning_rate": 0.0001868523746928479,
|
| 379 |
+
"loss": 1.0783,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.25364919837281646,
|
| 384 |
+
"grad_norm": 0.16735434292797727,
|
| 385 |
+
"learning_rate": 0.0001860111224655391,
|
| 386 |
+
"loss": 1.1149,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.25843503230437903,
|
| 391 |
+
"grad_norm": 0.1555718882454273,
|
| 392 |
+
"learning_rate": 0.0001851458015147673,
|
| 393 |
+
"loss": 1.1075,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2632208662359416,
|
| 398 |
+
"grad_norm": 0.16977397286003804,
|
| 399 |
+
"learning_rate": 0.00018425665398561883,
|
| 400 |
+
"loss": 1.0852,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.26800670016750416,
|
| 405 |
+
"grad_norm": 0.1663432558952086,
|
| 406 |
+
"learning_rate": 0.00018334392869063536,
|
| 407 |
+
"loss": 1.0811,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2727925340990668,
|
| 412 |
+
"grad_norm": 0.1666368907912767,
|
| 413 |
+
"learning_rate": 0.00018240788104018822,
|
| 414 |
+
"loss": 1.1014,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.27757836803062935,
|
| 419 |
+
"grad_norm": 0.15687924477714424,
|
| 420 |
+
"learning_rate": 0.00018144877297100606,
|
| 421 |
+
"loss": 1.0736,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2823642019621919,
|
| 426 |
+
"grad_norm": 0.16061101558707236,
|
| 427 |
+
"learning_rate": 0.0001804668728728764,
|
| 428 |
+
"loss": 1.0931,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2871500358937545,
|
| 433 |
+
"grad_norm": 0.160793101364396,
|
| 434 |
+
"learning_rate": 0.00017946245551354157,
|
| 435 |
+
"loss": 1.0999,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.29193586982531705,
|
| 440 |
+
"grad_norm": 0.1633263725476254,
|
| 441 |
+
"learning_rate": 0.00017843580196180952,
|
| 442 |
+
"loss": 1.0948,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2967217037568796,
|
| 447 |
+
"grad_norm": 0.1631452387362867,
|
| 448 |
+
"learning_rate": 0.00017738719950890168,
|
| 449 |
+
"loss": 1.1013,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3015075376884422,
|
| 454 |
+
"grad_norm": 0.16210535171429089,
|
| 455 |
+
"learning_rate": 0.00017631694158805946,
|
| 456 |
+
"loss": 1.0798,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.3062933716200048,
|
| 461 |
+
"grad_norm": 0.16229934636841442,
|
| 462 |
+
"learning_rate": 0.000175225327692432,
|
| 463 |
+
"loss": 1.0575,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.3110792055515674,
|
| 468 |
+
"grad_norm": 0.16588195121854765,
|
| 469 |
+
"learning_rate": 0.00017411266329126824,
|
| 470 |
+
"loss": 1.096,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.31586503948312994,
|
| 475 |
+
"grad_norm": 0.158210200244668,
|
| 476 |
+
"learning_rate": 0.00017297925974443673,
|
| 477 |
+
"loss": 1.1071,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3206508734146925,
|
| 482 |
+
"grad_norm": 0.1663784778299078,
|
| 483 |
+
"learning_rate": 0.00017182543421529676,
|
| 484 |
+
"loss": 1.0739,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.32543670734625507,
|
| 489 |
+
"grad_norm": 0.15384028200238806,
|
| 490 |
+
"learning_rate": 0.00017065150958194586,
|
| 491 |
+
"loss": 1.0848,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.33022254127781764,
|
| 496 |
+
"grad_norm": 0.16073214574887584,
|
| 497 |
+
"learning_rate": 0.00016945781434686783,
|
| 498 |
+
"loss": 1.1157,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.33500837520938026,
|
| 503 |
+
"grad_norm": 0.1745939193140414,
|
| 504 |
+
"learning_rate": 0.00016824468254500704,
|
| 505 |
+
"loss": 1.0815,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3397942091409428,
|
| 510 |
+
"grad_norm": 0.15802897019970708,
|
| 511 |
+
"learning_rate": 0.0001670124536502947,
|
| 512 |
+
"loss": 1.0779,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3445800430725054,
|
| 517 |
+
"grad_norm": 0.1579431494377225,
|
| 518 |
+
"learning_rate": 0.00016576147248065267,
|
| 519 |
+
"loss": 1.1031,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.34936587700406796,
|
| 524 |
+
"grad_norm": 0.16455288633589932,
|
| 525 |
+
"learning_rate": 0.00016449208910150232,
|
| 526 |
+
"loss": 1.1207,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3541517109356305,
|
| 531 |
+
"grad_norm": 0.15512720174775488,
|
| 532 |
+
"learning_rate": 0.00016320465872780477,
|
| 533 |
+
"loss": 1.0843,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3589375448671931,
|
| 538 |
+
"grad_norm": 0.15810739086397552,
|
| 539 |
+
"learning_rate": 0.00016189954162466012,
|
| 540 |
+
"loss": 1.0674,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.36372337879875566,
|
| 545 |
+
"grad_norm": 0.15539897008538223,
|
| 546 |
+
"learning_rate": 0.0001605771030064934,
|
| 547 |
+
"loss": 1.1075,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3685092127303183,
|
| 552 |
+
"grad_norm": 0.16059302879871643,
|
| 553 |
+
"learning_rate": 0.00015923771293485585,
|
| 554 |
+
"loss": 1.1083,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.37329504666188085,
|
| 559 |
+
"grad_norm": 0.1726863039386017,
|
| 560 |
+
"learning_rate": 0.00015788174621486934,
|
| 561 |
+
"loss": 1.0839,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3780808805934434,
|
| 566 |
+
"grad_norm": 0.160896911699282,
|
| 567 |
+
"learning_rate": 0.00015650958229034391,
|
| 568 |
+
"loss": 1.093,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.382866714525006,
|
| 573 |
+
"grad_norm": 0.1539033105501165,
|
| 574 |
+
"learning_rate": 0.00015512160513759672,
|
| 575 |
+
"loss": 1.0824,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.38765254845656855,
|
| 580 |
+
"grad_norm": 0.15253934847352404,
|
| 581 |
+
"learning_rate": 0.00015371820315800315,
|
| 582 |
+
"loss": 1.0611,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3924383823881311,
|
| 587 |
+
"grad_norm": 0.1549203336671571,
|
| 588 |
+
"learning_rate": 0.00015229976906930935,
|
| 589 |
+
"loss": 1.0926,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3972242163196937,
|
| 594 |
+
"grad_norm": 0.15736586699846142,
|
| 595 |
+
"learning_rate": 0.0001508666997957369,
|
| 596 |
+
"loss": 1.0838,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.4020100502512563,
|
| 601 |
+
"grad_norm": 0.15414651629074486,
|
| 602 |
+
"learning_rate": 0.00014941939635691035,
|
| 603 |
+
"loss": 1.0962,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.40679588418281887,
|
| 608 |
+
"grad_norm": 0.15216014768555902,
|
| 609 |
+
"learning_rate": 0.00014795826375563925,
|
| 610 |
+
"loss": 1.0837,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.41158171811438143,
|
| 615 |
+
"grad_norm": 0.1551252486012846,
|
| 616 |
+
"learning_rate": 0.0001464837108645845,
|
| 617 |
+
"loss": 1.096,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.416367552045944,
|
| 622 |
+
"grad_norm": 0.15880410911617168,
|
| 623 |
+
"learning_rate": 0.00014499615031184296,
|
| 624 |
+
"loss": 1.0947,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.42115338597750657,
|
| 629 |
+
"grad_norm": 0.16084656769756484,
|
| 630 |
+
"learning_rate": 0.00014349599836548034,
|
| 631 |
+
"loss": 1.0955,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.42593921990906913,
|
| 636 |
+
"grad_norm": 0.14942909791958908,
|
| 637 |
+
"learning_rate": 0.0001419836748170459,
|
| 638 |
+
"loss": 1.0911,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.43072505384063176,
|
| 643 |
+
"grad_norm": 0.16134597273400678,
|
| 644 |
+
"learning_rate": 0.0001404596028641009,
|
| 645 |
+
"loss": 1.1136,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4355108877721943,
|
| 650 |
+
"grad_norm": 0.15552785776756606,
|
| 651 |
+
"learning_rate": 0.0001389242089917943,
|
| 652 |
+
"loss": 1.1005,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4402967217037569,
|
| 657 |
+
"grad_norm": 0.1544583591443468,
|
| 658 |
+
"learning_rate": 0.00013737792285351805,
|
| 659 |
+
"loss": 1.0896,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.44508255563531945,
|
| 664 |
+
"grad_norm": 0.15743294110434283,
|
| 665 |
+
"learning_rate": 0.0001358211771506763,
|
| 666 |
+
"loss": 1.0687,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.449868389566882,
|
| 671 |
+
"grad_norm": 0.15489693015617015,
|
| 672 |
+
"learning_rate": 0.00013425440751160112,
|
| 673 |
+
"loss": 1.0909,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.4546542234984446,
|
| 678 |
+
"grad_norm": 0.1556280787651109,
|
| 679 |
+
"learning_rate": 0.00013267805236964967,
|
| 680 |
+
"loss": 1.1008,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.45944005743000715,
|
| 685 |
+
"grad_norm": 0.16139496091159036,
|
| 686 |
+
"learning_rate": 0.00013109255284051615,
|
| 687 |
+
"loss": 1.1167,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4642258913615698,
|
| 692 |
+
"grad_norm": 0.15380326887200926,
|
| 693 |
+
"learning_rate": 0.00012949835259879304,
|
| 694 |
+
"loss": 1.1021,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.46901172529313234,
|
| 699 |
+
"grad_norm": 0.1504710821626308,
|
| 700 |
+
"learning_rate": 0.00012789589775381676,
|
| 701 |
+
"loss": 1.0824,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4737975592246949,
|
| 706 |
+
"grad_norm": 0.16882632755621252,
|
| 707 |
+
"learning_rate": 0.00012628563672483146,
|
| 708 |
+
"loss": 1.091,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4785833931562575,
|
| 713 |
+
"grad_norm": 0.16236683430294702,
|
| 714 |
+
"learning_rate": 0.0001246680201155068,
|
| 715 |
+
"loss": 1.0609,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.48336922708782004,
|
| 720 |
+
"grad_norm": 0.1534881294655078,
|
| 721 |
+
"learning_rate": 0.00012304350058784405,
|
| 722 |
+
"loss": 1.0611,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.4881550610193826,
|
| 727 |
+
"grad_norm": 0.16620841316700394,
|
| 728 |
+
"learning_rate": 0.00012141253273550696,
|
| 729 |
+
"loss": 1.0932,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.49294089495094523,
|
| 734 |
+
"grad_norm": 0.16942714030828704,
|
| 735 |
+
"learning_rate": 0.00011977557295661108,
|
| 736 |
+
"loss": 1.0856,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4977267288825078,
|
| 741 |
+
"grad_norm": 0.15500201031703087,
|
| 742 |
+
"learning_rate": 0.00011813307932600887,
|
| 743 |
+
"loss": 1.0852,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5025125628140703,
|
| 748 |
+
"grad_norm": 0.15248801968172002,
|
| 749 |
+
"learning_rate": 0.00011648551146710556,
|
| 750 |
+
"loss": 1.1069,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5072983967456329,
|
| 755 |
+
"grad_norm": 0.14978453385390675,
|
| 756 |
+
"learning_rate": 0.0001148333304232411,
|
| 757 |
+
"loss": 1.088,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5120842306771956,
|
| 762 |
+
"grad_norm": 0.14736066147246124,
|
| 763 |
+
"learning_rate": 0.00011317699852867548,
|
| 764 |
+
"loss": 1.0506,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5168700646087581,
|
| 769 |
+
"grad_norm": 0.15088998664120562,
|
| 770 |
+
"learning_rate": 0.0001115169792792124,
|
| 771 |
+
"loss": 1.0972,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5216558985403207,
|
| 776 |
+
"grad_norm": 0.14676026138747209,
|
| 777 |
+
"learning_rate": 0.00010985373720249801,
|
| 778 |
+
"loss": 1.0871,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5264417324718832,
|
| 783 |
+
"grad_norm": 0.17054822297185676,
|
| 784 |
+
"learning_rate": 0.00010818773772803082,
|
| 785 |
+
"loss": 1.0957,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5312275664034458,
|
| 790 |
+
"grad_norm": 0.15081743477470166,
|
| 791 |
+
"learning_rate": 0.0001065194470569193,
|
| 792 |
+
"loss": 1.1114,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5360134003350083,
|
| 797 |
+
"grad_norm": 0.1556600989117304,
|
| 798 |
+
"learning_rate": 0.0001048493320314238,
|
| 799 |
+
"loss": 1.0747,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.540799234266571,
|
| 804 |
+
"grad_norm": 0.15346464585086714,
|
| 805 |
+
"learning_rate": 0.00010317786000431851,
|
| 806 |
+
"loss": 1.0761,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.5455850681981336,
|
| 811 |
+
"grad_norm": 0.15178562379014646,
|
| 812 |
+
"learning_rate": 0.00010150549870811107,
|
| 813 |
+
"loss": 1.0839,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5503709021296961,
|
| 818 |
+
"grad_norm": 0.15263581024104103,
|
| 819 |
+
"learning_rate": 9.983271612415575e-05,
|
| 820 |
+
"loss": 1.0742,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5551567360612587,
|
| 825 |
+
"grad_norm": 0.15166582071053056,
|
| 826 |
+
"learning_rate": 9.81599803516968e-05,
|
| 827 |
+
"loss": 1.0725,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5599425699928212,
|
| 832 |
+
"grad_norm": 0.14735687803417952,
|
| 833 |
+
"learning_rate": 9.648775947687912e-05,
|
| 834 |
+
"loss": 1.0705,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5647284039243838,
|
| 839 |
+
"grad_norm": 0.14825818203221888,
|
| 840 |
+
"learning_rate": 9.48165214417624e-05,
|
| 841 |
+
"loss": 1.0871,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5695142378559463,
|
| 846 |
+
"grad_norm": 0.15700946642781993,
|
| 847 |
+
"learning_rate": 9.314673391337576e-05,
|
| 848 |
+
"loss": 1.0979,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.574300071787509,
|
| 853 |
+
"grad_norm": 0.15580031067347558,
|
| 854 |
+
"learning_rate": 9.147886415284903e-05,
|
| 855 |
+
"loss": 1.0592,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5790859057190716,
|
| 860 |
+
"grad_norm": 0.14548002556094225,
|
| 861 |
+
"learning_rate": 8.981337888465788e-05,
|
| 862 |
+
"loss": 1.0787,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5838717396506341,
|
| 867 |
+
"grad_norm": 0.14237124600928142,
|
| 868 |
+
"learning_rate": 8.815074416601913e-05,
|
| 869 |
+
"loss": 1.0698,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5886575735821967,
|
| 874 |
+
"grad_norm": 0.15304745525626437,
|
| 875 |
+
"learning_rate": 8.649142525647272e-05,
|
| 876 |
+
"loss": 1.0848,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5934434075137592,
|
| 881 |
+
"grad_norm": 0.14513336716190856,
|
| 882 |
+
"learning_rate": 8.48358864876867e-05,
|
| 883 |
+
"loss": 1.0462,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5982292414453219,
|
| 888 |
+
"grad_norm": 0.1468415945819683,
|
| 889 |
+
"learning_rate": 8.318459113352221e-05,
|
| 890 |
+
"loss": 1.0906,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.6030150753768844,
|
| 895 |
+
"grad_norm": 0.14408143553897426,
|
| 896 |
+
"learning_rate": 8.153800128039441e-05,
|
| 897 |
+
"loss": 1.085,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.607800909308447,
|
| 902 |
+
"grad_norm": 0.15046217184291616,
|
| 903 |
+
"learning_rate": 7.989657769796533e-05,
|
| 904 |
+
"loss": 1.0882,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6125867432400096,
|
| 909 |
+
"grad_norm": 0.14348283659906289,
|
| 910 |
+
"learning_rate": 7.82607797102056e-05,
|
| 911 |
+
"loss": 1.0861,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6173725771715721,
|
| 916 |
+
"grad_norm": 0.14685503152106738,
|
| 917 |
+
"learning_rate": 7.663106506686057e-05,
|
| 918 |
+
"loss": 1.1003,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6221584111031347,
|
| 923 |
+
"grad_norm": 0.1480277391784376,
|
| 924 |
+
"learning_rate": 7.500788981535708e-05,
|
| 925 |
+
"loss": 1.0758,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6269442450346973,
|
| 930 |
+
"grad_norm": 0.1477910922274185,
|
| 931 |
+
"learning_rate": 7.339170817318625e-05,
|
| 932 |
+
"loss": 1.0695,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6317300789662599,
|
| 937 |
+
"grad_norm": 0.1551465349289344,
|
| 938 |
+
"learning_rate": 7.178297240079882e-05,
|
| 939 |
+
"loss": 1.0942,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.6365159128978225,
|
| 944 |
+
"grad_norm": 0.148811465121087,
|
| 945 |
+
"learning_rate": 7.018213267504775e-05,
|
| 946 |
+
"loss": 1.0825,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.641301746829385,
|
| 951 |
+
"grad_norm": 0.146937156337137,
|
| 952 |
+
"learning_rate": 6.858963696321403e-05,
|
| 953 |
+
"loss": 1.0985,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.6460875807609476,
|
| 958 |
+
"grad_norm": 0.14703161191479286,
|
| 959 |
+
"learning_rate": 6.700593089765086e-05,
|
| 960 |
+
"loss": 1.06,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6508734146925101,
|
| 965 |
+
"grad_norm": 0.14564360148371303,
|
| 966 |
+
"learning_rate": 6.543145765108106e-05,
|
| 967 |
+
"loss": 1.0853,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.6556592486240728,
|
| 972 |
+
"grad_norm": 0.14887365645849163,
|
| 973 |
+
"learning_rate": 6.3866657812583e-05,
|
| 974 |
+
"loss": 1.0787,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.6604450825556353,
|
| 979 |
+
"grad_norm": 0.14533659914404762,
|
| 980 |
+
"learning_rate": 6.231196926429913e-05,
|
| 981 |
+
"loss": 1.073,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.6652309164871979,
|
| 986 |
+
"grad_norm": 0.2354314895944445,
|
| 987 |
+
"learning_rate": 6.076782705890257e-05,
|
| 988 |
+
"loss": 1.0815,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6700167504187605,
|
| 993 |
+
"grad_norm": 0.14132233475416703,
|
| 994 |
+
"learning_rate": 5.9234663297854876e-05,
|
| 995 |
+
"loss": 1.0555,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.674802584350323,
|
| 1000 |
+
"grad_norm": 0.14913316600220797,
|
| 1001 |
+
"learning_rate": 5.7712907010490036e-05,
|
| 1002 |
+
"loss": 1.0785,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6795884182818857,
|
| 1007 |
+
"grad_norm": 0.15328072297180578,
|
| 1008 |
+
"learning_rate": 5.620298403395805e-05,
|
| 1009 |
+
"loss": 1.0857,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6843742522134482,
|
| 1014 |
+
"grad_norm": 0.17603388258774993,
|
| 1015 |
+
"learning_rate": 5.4705316894061765e-05,
|
| 1016 |
+
"loss": 1.0898,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6891600861450108,
|
| 1021 |
+
"grad_norm": 0.1448443355064005,
|
| 1022 |
+
"learning_rate": 5.322032468702036e-05,
|
| 1023 |
+
"loss": 1.0714,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6939459200765733,
|
| 1028 |
+
"grad_norm": 0.4624474555190123,
|
| 1029 |
+
"learning_rate": 5.1748422962192376e-05,
|
| 1030 |
+
"loss": 1.0994,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6987317540081359,
|
| 1035 |
+
"grad_norm": 0.14868980834848183,
|
| 1036 |
+
"learning_rate": 5.0290023605791666e-05,
|
| 1037 |
+
"loss": 1.0725,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.7035175879396985,
|
| 1042 |
+
"grad_norm": 0.15278504704361137,
|
| 1043 |
+
"learning_rate": 4.8845534725628086e-05,
|
| 1044 |
+
"loss": 1.0962,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.708303421871261,
|
| 1049 |
+
"grad_norm": 0.14605679246576617,
|
| 1050 |
+
"learning_rate": 4.741536053690552e-05,
|
| 1051 |
+
"loss": 1.0947,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7130892558028237,
|
| 1056 |
+
"grad_norm": 0.172204603811799,
|
| 1057 |
+
"learning_rate": 4.599990124910918e-05,
|
| 1058 |
+
"loss": 1.0758,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7178750897343862,
|
| 1063 |
+
"grad_norm": 0.14357849865669614,
|
| 1064 |
+
"learning_rate": 4.4599552954014145e-05,
|
| 1065 |
+
"loss": 1.0682,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7226609236659488,
|
| 1070 |
+
"grad_norm": 0.14980923833672957,
|
| 1071 |
+
"learning_rate": 4.32147075148458e-05,
|
| 1072 |
+
"loss": 1.0814,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7274467575975113,
|
| 1077 |
+
"grad_norm": 0.16395768222951593,
|
| 1078 |
+
"learning_rate": 4.1845752456623665e-05,
|
| 1079 |
+
"loss": 1.0583,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.7322325915290739,
|
| 1084 |
+
"grad_norm": 0.14059821304657993,
|
| 1085 |
+
"learning_rate": 4.049307085771931e-05,
|
| 1086 |
+
"loss": 1.0839,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.7370184254606366,
|
| 1091 |
+
"grad_norm": 0.1472110334031576,
|
| 1092 |
+
"learning_rate": 3.9157041242658477e-05,
|
| 1093 |
+
"loss": 1.1079,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.7418042593921991,
|
| 1098 |
+
"grad_norm": 0.14020342123522012,
|
| 1099 |
+
"learning_rate": 3.783803747619741e-05,
|
| 1100 |
+
"loss": 1.0829,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.7465900933237617,
|
| 1105 |
+
"grad_norm": 0.17437047699695307,
|
| 1106 |
+
"learning_rate": 3.653642865870359e-05,
|
| 1107 |
+
"loss": 1.0808,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.7513759272553242,
|
| 1112 |
+
"grad_norm": 0.14320013892049976,
|
| 1113 |
+
"learning_rate": 3.525257902286908e-05,
|
| 1114 |
+
"loss": 1.0608,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.7561617611868868,
|
| 1119 |
+
"grad_norm": 0.14437417000631428,
|
| 1120 |
+
"learning_rate": 3.398684783178648e-05,
|
| 1121 |
+
"loss": 1.0618,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.7609475951184493,
|
| 1126 |
+
"grad_norm": 0.14321363672597254,
|
| 1127 |
+
"learning_rate": 3.273958927841525e-05,
|
| 1128 |
+
"loss": 1.0659,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.765733429050012,
|
| 1133 |
+
"grad_norm": 0.14121990349576288,
|
| 1134 |
+
"learning_rate": 3.1511152386467055e-05,
|
| 1135 |
+
"loss": 1.0936,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.7705192629815746,
|
| 1140 |
+
"grad_norm": 0.16146069783583863,
|
| 1141 |
+
"learning_rate": 3.0301880912737568e-05,
|
| 1142 |
+
"loss": 1.0647,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.7753050969131371,
|
| 1147 |
+
"grad_norm": 0.1447026626027737,
|
| 1148 |
+
"learning_rate": 2.9112113250911844e-05,
|
| 1149 |
+
"loss": 1.0747,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.7800909308446997,
|
| 1154 |
+
"grad_norm": 0.14724228311552523,
|
| 1155 |
+
"learning_rate": 2.7942182336870925e-05,
|
| 1156 |
+
"loss": 1.1046,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.7848767647762622,
|
| 1161 |
+
"grad_norm": 0.14612792897080507,
|
| 1162 |
+
"learning_rate": 2.6792415555525463e-05,
|
| 1163 |
+
"loss": 1.0391,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.7896625987078248,
|
| 1168 |
+
"grad_norm": 0.14445016139434405,
|
| 1169 |
+
"learning_rate": 2.5663134649202647e-05,
|
| 1170 |
+
"loss": 1.0808,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.7944484326393874,
|
| 1175 |
+
"grad_norm": 0.14283033243615206,
|
| 1176 |
+
"learning_rate": 2.4554655627612245e-05,
|
| 1177 |
+
"loss": 1.0767,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.79923426657095,
|
| 1182 |
+
"grad_norm": 0.1428104588189023,
|
| 1183 |
+
"learning_rate": 2.34672886794167e-05,
|
| 1184 |
+
"loss": 1.0884,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.8040201005025126,
|
| 1189 |
+
"grad_norm": 0.14106416222944104,
|
| 1190 |
+
"learning_rate": 2.2401338085430323e-05,
|
| 1191 |
+
"loss": 1.0891,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.8088059344340751,
|
| 1196 |
+
"grad_norm": 0.14453431354715718,
|
| 1197 |
+
"learning_rate": 2.135710213347134e-05,
|
| 1198 |
+
"loss": 1.0829,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.8135917683656377,
|
| 1203 |
+
"grad_norm": 0.1436138017414945,
|
| 1204 |
+
"learning_rate": 2.0334873034891554e-05,
|
| 1205 |
+
"loss": 1.0823,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.8183776022972002,
|
| 1210 |
+
"grad_norm": 0.14415504753616376,
|
| 1211 |
+
"learning_rate": 1.933493684280574e-05,
|
| 1212 |
+
"loss": 1.0749,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8231634362287629,
|
| 1217 |
+
"grad_norm": 0.14188286670890893,
|
| 1218 |
+
"learning_rate": 1.8357573372044834e-05,
|
| 1219 |
+
"loss": 1.0775,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.8279492701603255,
|
| 1224 |
+
"grad_norm": 0.14043422592547342,
|
| 1225 |
+
"learning_rate": 1.740305612085439e-05,
|
| 1226 |
+
"loss": 1.0852,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.832735104091888,
|
| 1231 |
+
"grad_norm": 0.14014109535516273,
|
| 1232 |
+
"learning_rate": 1.647165219436113e-05,
|
| 1233 |
+
"loss": 1.0716,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.8375209380234506,
|
| 1238 |
+
"grad_norm": 0.18266681120520475,
|
| 1239 |
+
"learning_rate": 1.556362222982799e-05,
|
| 1240 |
+
"loss": 1.0711,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.8423067719550131,
|
| 1245 |
+
"grad_norm": 0.14585487433506303,
|
| 1246 |
+
"learning_rate": 1.4679220323719234e-05,
|
| 1247 |
+
"loss": 1.0561,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.8470926058865758,
|
| 1252 |
+
"grad_norm": 0.13911103035630754,
|
| 1253 |
+
"learning_rate": 1.3818693960596185e-05,
|
| 1254 |
+
"loss": 1.0707,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.8518784398181383,
|
| 1259 |
+
"grad_norm": 0.15612123605821987,
|
| 1260 |
+
"learning_rate": 1.2982283943862738e-05,
|
| 1261 |
+
"loss": 1.0494,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.8566642737497009,
|
| 1266 |
+
"grad_norm": 0.14067555622023134,
|
| 1267 |
+
"learning_rate": 1.217022432838093e-05,
|
| 1268 |
+
"loss": 1.0686,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.8614501076812635,
|
| 1273 |
+
"grad_norm": 0.1457410414761679,
|
| 1274 |
+
"learning_rate": 1.1382742354974429e-05,
|
| 1275 |
+
"loss": 1.0562,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.866235941612826,
|
| 1280 |
+
"grad_norm": 0.1398250627278749,
|
| 1281 |
+
"learning_rate": 1.0620058386839393e-05,
|
| 1282 |
+
"loss": 1.0753,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.8710217755443886,
|
| 1287 |
+
"grad_norm": 0.14690238478434312,
|
| 1288 |
+
"learning_rate": 9.882385847879539e-06,
|
| 1289 |
+
"loss": 1.0539,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.8758076094759512,
|
| 1294 |
+
"grad_norm": 0.14224902345010998,
|
| 1295 |
+
"learning_rate": 9.169931162983137e-06,
|
| 1296 |
+
"loss": 1.0575,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.8805934434075138,
|
| 1301 |
+
"grad_norm": 0.14002967562121116,
|
| 1302 |
+
"learning_rate": 8.482893700258643e-06,
|
| 1303 |
+
"loss": 1.0831,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.8853792773390763,
|
| 1308 |
+
"grad_norm": 0.14652920530592364,
|
| 1309 |
+
"learning_rate": 7.821465715244947e-06,
|
| 1310 |
+
"loss": 1.0844,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.8901651112706389,
|
| 1315 |
+
"grad_norm": 0.13985808750925746,
|
| 1316 |
+
"learning_rate": 7.185832297111938e-06,
|
| 1317 |
+
"loss": 1.0618,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.8949509452022015,
|
| 1322 |
+
"grad_norm": 0.15160308510490375,
|
| 1323 |
+
"learning_rate": 6.576171316866608e-06,
|
| 1324 |
+
"loss": 1.0773,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.899736779133764,
|
| 1329 |
+
"grad_norm": 0.14784429409642344,
|
| 1330 |
+
"learning_rate": 5.9926533775789055e-06,
|
| 1331 |
+
"loss": 1.0951,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.9045226130653267,
|
| 1336 |
+
"grad_norm": 0.14167088318411009,
|
| 1337 |
+
"learning_rate": 5.435441766641369e-06,
|
| 1338 |
+
"loss": 1.0841,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.9093084469968892,
|
| 1343 |
+
"grad_norm": 0.14256818695069146,
|
| 1344 |
+
"learning_rate": 4.904692410075973e-06,
|
| 1345 |
+
"loss": 1.0647,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.9140942809284518,
|
| 1350 |
+
"grad_norm": 0.15531748633710526,
|
| 1351 |
+
"learning_rate": 4.400553828900989e-06,
|
| 1352 |
+
"loss": 1.0757,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.9188801148600143,
|
| 1357 |
+
"grad_norm": 0.14420681549864126,
|
| 1358 |
+
"learning_rate": 3.923167097569935e-06,
|
| 1359 |
+
"loss": 1.0903,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.9236659487915769,
|
| 1364 |
+
"grad_norm": 0.14398010788396462,
|
| 1365 |
+
"learning_rate": 3.4726658044943126e-06,
|
| 1366 |
+
"loss": 1.0668,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.9284517827231396,
|
| 1371 |
+
"grad_norm": 0.14589900176146645,
|
| 1372 |
+
"learning_rate": 3.0491760146611926e-06,
|
| 1373 |
+
"loss": 1.0845,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.9332376166547021,
|
| 1378 |
+
"grad_norm": 0.13882750982702796,
|
| 1379 |
+
"learning_rate": 2.652816234356159e-06,
|
| 1380 |
+
"loss": 1.0382,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.9380234505862647,
|
| 1385 |
+
"grad_norm": 0.14112035905216325,
|
| 1386 |
+
"learning_rate": 2.283697378001315e-06,
|
| 1387 |
+
"loss": 1.0825,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.9428092845178272,
|
| 1392 |
+
"grad_norm": 0.13934480624047157,
|
| 1393 |
+
"learning_rate": 1.9419227371178627e-06,
|
| 1394 |
+
"loss": 1.0679,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.9475951184493898,
|
| 1399 |
+
"grad_norm": 0.14117739445269173,
|
| 1400 |
+
"learning_rate": 1.6275879514217052e-06,
|
| 1401 |
+
"loss": 1.0772,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.9523809523809523,
|
| 1406 |
+
"grad_norm": 0.14031209854381504,
|
| 1407 |
+
"learning_rate": 1.3407809820603856e-06,
|
| 1408 |
+
"loss": 1.0767,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.957166786312515,
|
| 1413 |
+
"grad_norm": 0.14091355128035063,
|
| 1414 |
+
"learning_rate": 1.0815820869985893e-06,
|
| 1415 |
+
"loss": 1.0635,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.9619526202440776,
|
| 1420 |
+
"grad_norm": 0.14100298765660577,
|
| 1421 |
+
"learning_rate": 8.50063798559475e-07,
|
| 1422 |
+
"loss": 1.0861,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.9667384541756401,
|
| 1427 |
+
"grad_norm": 0.1412224316948679,
|
| 1428 |
+
"learning_rate": 6.462909031276443e-07,
|
| 1429 |
+
"loss": 1.0633,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.9715242881072027,
|
| 1434 |
+
"grad_norm": 0.1385906964183353,
|
| 1435 |
+
"learning_rate": 4.7032042301985434e-07,
|
| 1436 |
+
"loss": 1.0726,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.9763101220387652,
|
| 1441 |
+
"grad_norm": 0.151976410727331,
|
| 1442 |
+
"learning_rate": 3.222016005282824e-07,
|
| 1443 |
+
"loss": 1.0645,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.9810959559703278,
|
| 1448 |
+
"grad_norm": 0.14304003914264313,
|
| 1449 |
+
"learning_rate": 2.0197588414094804e-07,
|
| 1450 |
+
"loss": 1.0785,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.9858817899018905,
|
| 1455 |
+
"grad_norm": 0.1395525833943131,
|
| 1456 |
+
"learning_rate": 1.0967691694302451e-07,
|
| 1457 |
+
"loss": 1.0582,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.990667623833453,
|
| 1462 |
+
"grad_norm": 0.14208441339069042,
|
| 1463 |
+
"learning_rate": 4.5330527202480654e-08,
|
| 1464 |
+
"loss": 1.0763,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.9954534577650156,
|
| 1469 |
+
"grad_norm": 0.13931375530799342,
|
| 1470 |
+
"learning_rate": 8.95472114241791e-09,
|
| 1471 |
+
"loss": 1.0444,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.9992821249102656,
|
| 1476 |
+
"eval_loss": 1.077100157737732,
|
| 1477 |
+
"eval_runtime": 3923.6787,
|
| 1478 |
+
"eval_samples_per_second": 3.43,
|
| 1479 |
+
"eval_steps_per_second": 0.858,
|
| 1480 |
+
"step": 1044
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.9992821249102656,
|
| 1484 |
+
"step": 1044,
|
| 1485 |
+
"total_flos": 2155604625850368.0,
|
| 1486 |
+
"train_loss": 1.091635976486279,
|
| 1487 |
+
"train_runtime": 24351.9745,
|
| 1488 |
+
"train_samples_per_second": 2.746,
|
| 1489 |
+
"train_steps_per_second": 0.043
|
| 1490 |
+
}
|
| 1491 |
+
],
|
| 1492 |
+
"logging_steps": 5,
|
| 1493 |
+
"max_steps": 1044,
|
| 1494 |
+
"num_input_tokens_seen": 0,
|
| 1495 |
+
"num_train_epochs": 1,
|
| 1496 |
+
"save_steps": 100,
|
| 1497 |
+
"stateful_callbacks": {
|
| 1498 |
+
"TrainerControl": {
|
| 1499 |
+
"args": {
|
| 1500 |
+
"should_epoch_stop": false,
|
| 1501 |
+
"should_evaluate": false,
|
| 1502 |
+
"should_log": false,
|
| 1503 |
+
"should_save": true,
|
| 1504 |
+
"should_training_stop": true
|
| 1505 |
+
},
|
| 1506 |
+
"attributes": {}
|
| 1507 |
+
}
|
| 1508 |
+
},
|
| 1509 |
+
"total_flos": 2155604625850368.0,
|
| 1510 |
+
"train_batch_size": 4,
|
| 1511 |
+
"trial_name": null,
|
| 1512 |
+
"trial_params": null
|
| 1513 |
+
}
|