Model save
Browse files- README.md +69 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1394 -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_0.02p_seed42_level2_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_0.02p_seed42_level2_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: 1.1101
|
| 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 |
+
| 1.071 | 1.0 | 959 | 1.1101 |
|
| 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": 1.0,
|
| 3 |
+
"total_flos": 1.2060020938113024e+16,
|
| 4 |
+
"train_loss": 1.1071907986689657,
|
| 5 |
+
"train_runtime": 19793.4474,
|
| 6 |
+
"train_samples": 104971,
|
| 7 |
+
"train_samples_per_second": 3.101,
|
| 8 |
+
"train_steps_per_second": 0.048
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.2060020938113024e+16,
|
| 4 |
+
"train_loss": 1.1071907986689657,
|
| 5 |
+
"train_runtime": 19793.4474,
|
| 6 |
+
"train_samples": 104971,
|
| 7 |
+
"train_samples_per_second": 3.101,
|
| 8 |
+
"train_steps_per_second": 0.048
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1394 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 959,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0010427528675703858,
|
| 13 |
+
"grad_norm": 0.46784728546447485,
|
| 14 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 15 |
+
"loss": 1.3538,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005213764337851929,
|
| 20 |
+
"grad_norm": 0.4299186002140457,
|
| 21 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 22 |
+
"loss": 1.3601,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010427528675703858,
|
| 27 |
+
"grad_norm": 0.5025473100643405,
|
| 28 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 29 |
+
"loss": 1.3602,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01564129301355579,
|
| 34 |
+
"grad_norm": 0.22754157012043386,
|
| 35 |
+
"learning_rate": 3.125e-05,
|
| 36 |
+
"loss": 1.3526,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.020855057351407715,
|
| 41 |
+
"grad_norm": 0.14930678513355825,
|
| 42 |
+
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 1.3157,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.026068821689259645,
|
| 48 |
+
"grad_norm": 0.1493862762381592,
|
| 49 |
+
"learning_rate": 5.208333333333334e-05,
|
| 50 |
+
"loss": 1.2959,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.03128258602711158,
|
| 55 |
+
"grad_norm": 0.14284419715578545,
|
| 56 |
+
"learning_rate": 6.25e-05,
|
| 57 |
+
"loss": 1.2774,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0364963503649635,
|
| 62 |
+
"grad_norm": 0.13626355620208974,
|
| 63 |
+
"learning_rate": 7.291666666666667e-05,
|
| 64 |
+
"loss": 1.2487,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04171011470281543,
|
| 69 |
+
"grad_norm": 0.10695646364370354,
|
| 70 |
+
"learning_rate": 8.333333333333334e-05,
|
| 71 |
+
"loss": 1.2206,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04692387904066736,
|
| 76 |
+
"grad_norm": 0.09407825046309691,
|
| 77 |
+
"learning_rate": 9.375e-05,
|
| 78 |
+
"loss": 1.2028,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.05213764337851929,
|
| 83 |
+
"grad_norm": 0.08120415093523357,
|
| 84 |
+
"learning_rate": 0.00010416666666666667,
|
| 85 |
+
"loss": 1.1819,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05735140771637122,
|
| 90 |
+
"grad_norm": 0.07854484960093237,
|
| 91 |
+
"learning_rate": 0.00011458333333333333,
|
| 92 |
+
"loss": 1.1687,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06256517205422316,
|
| 97 |
+
"grad_norm": 0.06991743542761439,
|
| 98 |
+
"learning_rate": 0.000125,
|
| 99 |
+
"loss": 1.1476,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.06777893639207508,
|
| 104 |
+
"grad_norm": 0.07010747792123796,
|
| 105 |
+
"learning_rate": 0.0001354166666666667,
|
| 106 |
+
"loss": 1.1566,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.072992700729927,
|
| 111 |
+
"grad_norm": 0.06735485294531884,
|
| 112 |
+
"learning_rate": 0.00014583333333333335,
|
| 113 |
+
"loss": 1.1758,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.07820646506777894,
|
| 118 |
+
"grad_norm": 0.07054279022592949,
|
| 119 |
+
"learning_rate": 0.00015625,
|
| 120 |
+
"loss": 1.1725,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08342022940563086,
|
| 125 |
+
"grad_norm": 0.07848476612174879,
|
| 126 |
+
"learning_rate": 0.0001666666666666667,
|
| 127 |
+
"loss": 1.1588,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0886339937434828,
|
| 132 |
+
"grad_norm": 0.07428124015010305,
|
| 133 |
+
"learning_rate": 0.00017708333333333335,
|
| 134 |
+
"loss": 1.1409,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.09384775808133472,
|
| 139 |
+
"grad_norm": 0.07511025896318015,
|
| 140 |
+
"learning_rate": 0.0001875,
|
| 141 |
+
"loss": 1.1501,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.09906152241918666,
|
| 146 |
+
"grad_norm": 0.08140724060317898,
|
| 147 |
+
"learning_rate": 0.0001979166666666667,
|
| 148 |
+
"loss": 1.1525,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.10427528675703858,
|
| 153 |
+
"grad_norm": 0.09875089257447961,
|
| 154 |
+
"learning_rate": 0.00019998939866722242,
|
| 155 |
+
"loss": 1.1152,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.10948905109489052,
|
| 160 |
+
"grad_norm": 0.10683544050076133,
|
| 161 |
+
"learning_rate": 0.00019994633460515538,
|
| 162 |
+
"loss": 1.1245,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.11470281543274244,
|
| 167 |
+
"grad_norm": 0.10489956661825363,
|
| 168 |
+
"learning_rate": 0.00019987015948614691,
|
| 169 |
+
"loss": 1.1228,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.11991657977059438,
|
| 174 |
+
"grad_norm": 0.07725971069252748,
|
| 175 |
+
"learning_rate": 0.00019976089854612702,
|
| 176 |
+
"loss": 1.1215,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.1251303441084463,
|
| 181 |
+
"grad_norm": 0.07944990580241551,
|
| 182 |
+
"learning_rate": 0.000199618587981973,
|
| 183 |
+
"loss": 1.116,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13034410844629823,
|
| 188 |
+
"grad_norm": 0.0834401702395657,
|
| 189 |
+
"learning_rate": 0.00019944327493951773,
|
| 190 |
+
"loss": 1.1144,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.13555787278415016,
|
| 195 |
+
"grad_norm": 0.08900854650193943,
|
| 196 |
+
"learning_rate": 0.00019923501749793097,
|
| 197 |
+
"loss": 1.1048,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.14077163712200208,
|
| 202 |
+
"grad_norm": 0.08078088093283514,
|
| 203 |
+
"learning_rate": 0.0001989938846504783,
|
| 204 |
+
"loss": 1.1022,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.145985401459854,
|
| 209 |
+
"grad_norm": 0.08120650571343319,
|
| 210 |
+
"learning_rate": 0.00019871995628166443,
|
| 211 |
+
"loss": 1.1269,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.15119916579770595,
|
| 216 |
+
"grad_norm": 0.08251173811384645,
|
| 217 |
+
"learning_rate": 0.00019841332314076855,
|
| 218 |
+
"loss": 1.1049,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.15641293013555788,
|
| 223 |
+
"grad_norm": 0.0811645274318103,
|
| 224 |
+
"learning_rate": 0.00019807408681177997,
|
| 225 |
+
"loss": 1.1132,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.1616266944734098,
|
| 230 |
+
"grad_norm": 0.07646936117341427,
|
| 231 |
+
"learning_rate": 0.00019770235967974463,
|
| 232 |
+
"loss": 1.1058,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.16684045881126172,
|
| 237 |
+
"grad_norm": 0.07385173118278343,
|
| 238 |
+
"learning_rate": 0.00019729826489353324,
|
| 239 |
+
"loss": 1.1209,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.17205422314911367,
|
| 244 |
+
"grad_norm": 0.08021452641715054,
|
| 245 |
+
"learning_rate": 0.00019686193632504338,
|
| 246 |
+
"loss": 1.1049,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1772679874869656,
|
| 251 |
+
"grad_norm": 0.07668807421928069,
|
| 252 |
+
"learning_rate": 0.00019639351852484947,
|
| 253 |
+
"loss": 1.1106,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.18248175182481752,
|
| 258 |
+
"grad_norm": 0.07892223313284867,
|
| 259 |
+
"learning_rate": 0.0001958931666743148,
|
| 260 |
+
"loss": 1.1161,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.18769551616266944,
|
| 265 |
+
"grad_norm": 0.08190647698912198,
|
| 266 |
+
"learning_rate": 0.0001953610465341816,
|
| 267 |
+
"loss": 1.0996,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.19290928050052136,
|
| 272 |
+
"grad_norm": 0.07800476181033798,
|
| 273 |
+
"learning_rate": 0.00019479733438965667,
|
| 274 |
+
"loss": 1.1247,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1981230448383733,
|
| 279 |
+
"grad_norm": 0.08349049942704777,
|
| 280 |
+
"learning_rate": 0.00019420221699200995,
|
| 281 |
+
"loss": 1.1154,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.20333680917622524,
|
| 286 |
+
"grad_norm": 0.07929575871932729,
|
| 287 |
+
"learning_rate": 0.00019357589149670608,
|
| 288 |
+
"loss": 1.1184,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.20855057351407716,
|
| 293 |
+
"grad_norm": 0.0735399192652975,
|
| 294 |
+
"learning_rate": 0.00019291856539808917,
|
| 295 |
+
"loss": 1.0968,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.21376433785192908,
|
| 300 |
+
"grad_norm": 0.07775467151795544,
|
| 301 |
+
"learning_rate": 0.00019223045646064212,
|
| 302 |
+
"loss": 1.1122,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.21897810218978103,
|
| 307 |
+
"grad_norm": 0.07608187355699407,
|
| 308 |
+
"learning_rate": 0.00019151179264684402,
|
| 309 |
+
"loss": 1.1129,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.22419186652763295,
|
| 314 |
+
"grad_norm": 0.07668601971351766,
|
| 315 |
+
"learning_rate": 0.00019076281204164874,
|
| 316 |
+
"loss": 1.0979,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.22940563086548488,
|
| 321 |
+
"grad_norm": 0.07230005483681756,
|
| 322 |
+
"learning_rate": 0.00018998376277361008,
|
| 323 |
+
"loss": 1.1215,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2346193952033368,
|
| 328 |
+
"grad_norm": 0.0733672154981931,
|
| 329 |
+
"learning_rate": 0.00018917490293267973,
|
| 330 |
+
"loss": 1.1021,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.23983315954118875,
|
| 335 |
+
"grad_norm": 0.08070384712102877,
|
| 336 |
+
"learning_rate": 0.00018833650048470524,
|
| 337 |
+
"loss": 1.1131,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.24504692387904067,
|
| 342 |
+
"grad_norm": 0.0789014320960882,
|
| 343 |
+
"learning_rate": 0.0001874688331826557,
|
| 344 |
+
"loss": 1.1226,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.2502606882168926,
|
| 349 |
+
"grad_norm": 0.0713759955633646,
|
| 350 |
+
"learning_rate": 0.00018657218847460572,
|
| 351 |
+
"loss": 1.124,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.25547445255474455,
|
| 356 |
+
"grad_norm": 0.07275793106627063,
|
| 357 |
+
"learning_rate": 0.00018564686340850708,
|
| 358 |
+
"loss": 1.1081,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.26068821689259647,
|
| 363 |
+
"grad_norm": 0.07708474011231153,
|
| 364 |
+
"learning_rate": 0.0001846931645337803,
|
| 365 |
+
"loss": 1.094,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2659019812304484,
|
| 370 |
+
"grad_norm": 0.08547989244439409,
|
| 371 |
+
"learning_rate": 0.00018371140779975824,
|
| 372 |
+
"loss": 1.1068,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2711157455683003,
|
| 377 |
+
"grad_norm": 0.07980805306815425,
|
| 378 |
+
"learning_rate": 0.00018270191845101602,
|
| 379 |
+
"loss": 1.1097,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.27632950990615224,
|
| 384 |
+
"grad_norm": 0.0771809814717238,
|
| 385 |
+
"learning_rate": 0.0001816650309196209,
|
| 386 |
+
"loss": 1.0993,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.28154327424400416,
|
| 391 |
+
"grad_norm": 0.07825602560329317,
|
| 392 |
+
"learning_rate": 0.00018060108871433922,
|
| 393 |
+
"loss": 1.0965,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2867570385818561,
|
| 398 |
+
"grad_norm": 0.07544233821374238,
|
| 399 |
+
"learning_rate": 0.00017951044430683565,
|
| 400 |
+
"loss": 1.1077,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.291970802919708,
|
| 405 |
+
"grad_norm": 0.08595174428600973,
|
| 406 |
+
"learning_rate": 0.00017839345901490367,
|
| 407 |
+
"loss": 1.1444,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.29718456725756,
|
| 412 |
+
"grad_norm": 0.08036998457602522,
|
| 413 |
+
"learning_rate": 0.000177250502882765,
|
| 414 |
+
"loss": 1.1018,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.3023983315954119,
|
| 419 |
+
"grad_norm": 0.0758408964978662,
|
| 420 |
+
"learning_rate": 0.0001760819545584783,
|
| 421 |
+
"loss": 1.1144,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.30761209593326383,
|
| 426 |
+
"grad_norm": 0.08039719317248988,
|
| 427 |
+
"learning_rate": 0.00017488820116849757,
|
| 428 |
+
"loss": 1.0848,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.31282586027111575,
|
| 433 |
+
"grad_norm": 0.08195730354878057,
|
| 434 |
+
"learning_rate": 0.0001736696381894216,
|
| 435 |
+
"loss": 1.1091,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3180396246089677,
|
| 440 |
+
"grad_norm": 0.0857915617590216,
|
| 441 |
+
"learning_rate": 0.0001724266693169772,
|
| 442 |
+
"loss": 1.1179,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.3232533889468196,
|
| 447 |
+
"grad_norm": 0.0727205241081358,
|
| 448 |
+
"learning_rate": 0.00017115970633227936,
|
| 449 |
+
"loss": 1.1104,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3284671532846715,
|
| 454 |
+
"grad_norm": 0.08177574294024298,
|
| 455 |
+
"learning_rate": 0.00016986916896541323,
|
| 456 |
+
"loss": 1.1,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.33368091762252344,
|
| 461 |
+
"grad_norm": 0.07784461413056201,
|
| 462 |
+
"learning_rate": 0.00016855548475638225,
|
| 463 |
+
"loss": 1.0934,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.33889468196037537,
|
| 468 |
+
"grad_norm": 0.0708482833267521,
|
| 469 |
+
"learning_rate": 0.0001672190889134691,
|
| 470 |
+
"loss": 1.0958,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.34410844629822734,
|
| 475 |
+
"grad_norm": 0.076390205474366,
|
| 476 |
+
"learning_rate": 0.0001658604241690564,
|
| 477 |
+
"loss": 1.105,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.34932221063607927,
|
| 482 |
+
"grad_norm": 0.07309708286911666,
|
| 483 |
+
"learning_rate": 0.00016447994063295457,
|
| 484 |
+
"loss": 1.095,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3545359749739312,
|
| 489 |
+
"grad_norm": 0.07017926439528561,
|
| 490 |
+
"learning_rate": 0.00016307809564328551,
|
| 491 |
+
"loss": 1.1157,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.3597497393117831,
|
| 496 |
+
"grad_norm": 0.07365098763599127,
|
| 497 |
+
"learning_rate": 0.00016165535361497218,
|
| 498 |
+
"loss": 1.1186,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.36496350364963503,
|
| 503 |
+
"grad_norm": 0.07663605771628923,
|
| 504 |
+
"learning_rate": 0.00016021218588588298,
|
| 505 |
+
"loss": 1.0749,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.37017726798748696,
|
| 510 |
+
"grad_norm": 0.07615854076662545,
|
| 511 |
+
"learning_rate": 0.0001587490705606832,
|
| 512 |
+
"loss": 1.0827,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3753910323253389,
|
| 517 |
+
"grad_norm": 0.0753345353309753,
|
| 518 |
+
"learning_rate": 0.00015726649235244455,
|
| 519 |
+
"loss": 1.1082,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3806047966631908,
|
| 524 |
+
"grad_norm": 0.07913485493599222,
|
| 525 |
+
"learning_rate": 0.00015576494242206508,
|
| 526 |
+
"loss": 1.1258,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3858185610010427,
|
| 531 |
+
"grad_norm": 0.07705620467095986,
|
| 532 |
+
"learning_rate": 0.0001542449182155537,
|
| 533 |
+
"loss": 1.0772,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3910323253388947,
|
| 538 |
+
"grad_norm": 0.07160314888274456,
|
| 539 |
+
"learning_rate": 0.00015270692329923176,
|
| 540 |
+
"loss": 1.0996,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3962460896767466,
|
| 545 |
+
"grad_norm": 0.08325118544541214,
|
| 546 |
+
"learning_rate": 0.0001511514671929076,
|
| 547 |
+
"loss": 1.0904,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.40145985401459855,
|
| 552 |
+
"grad_norm": 0.07199438515468881,
|
| 553 |
+
"learning_rate": 0.00014957906520107845,
|
| 554 |
+
"loss": 1.0851,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.40667361835245047,
|
| 559 |
+
"grad_norm": 0.07710028013752196,
|
| 560 |
+
"learning_rate": 0.00014799023824221613,
|
| 561 |
+
"loss": 1.1123,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4118873826903024,
|
| 566 |
+
"grad_norm": 0.07407414298584401,
|
| 567 |
+
"learning_rate": 0.0001463855126761928,
|
| 568 |
+
"loss": 1.0878,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4171011470281543,
|
| 573 |
+
"grad_norm": 0.07467701529997763,
|
| 574 |
+
"learning_rate": 0.0001447654201299041,
|
| 575 |
+
"loss": 1.0975,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.42231491136600624,
|
| 580 |
+
"grad_norm": 0.07107152720346374,
|
| 581 |
+
"learning_rate": 0.00014313049732114715,
|
| 582 |
+
"loss": 1.0657,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.42752867570385816,
|
| 587 |
+
"grad_norm": 0.07823458012876569,
|
| 588 |
+
"learning_rate": 0.00014148128588081256,
|
| 589 |
+
"loss": 1.1153,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.43274244004171014,
|
| 594 |
+
"grad_norm": 0.08009223556489412,
|
| 595 |
+
"learning_rate": 0.0001398183321734481,
|
| 596 |
+
"loss": 1.0974,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.43795620437956206,
|
| 601 |
+
"grad_norm": 0.07669197698500169,
|
| 602 |
+
"learning_rate": 0.00013814218711625457,
|
| 603 |
+
"loss": 1.0904,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.443169968717414,
|
| 608 |
+
"grad_norm": 0.08468765745640104,
|
| 609 |
+
"learning_rate": 0.0001364534059965735,
|
| 610 |
+
"loss": 1.0808,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.4483837330552659,
|
| 615 |
+
"grad_norm": 0.07933899715254406,
|
| 616 |
+
"learning_rate": 0.00013475254828792662,
|
| 617 |
+
"loss": 1.1232,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.45359749739311783,
|
| 622 |
+
"grad_norm": 0.07992483417151192,
|
| 623 |
+
"learning_rate": 0.0001330401774646691,
|
| 624 |
+
"loss": 1.0752,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.45881126173096975,
|
| 629 |
+
"grad_norm": 0.07194633805909506,
|
| 630 |
+
"learning_rate": 0.00013131686081531698,
|
| 631 |
+
"loss": 1.1021,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4640250260688217,
|
| 636 |
+
"grad_norm": 0.07408216210208647,
|
| 637 |
+
"learning_rate": 0.00012958316925461085,
|
| 638 |
+
"loss": 1.103,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.4692387904066736,
|
| 643 |
+
"grad_norm": 0.07365408906170111,
|
| 644 |
+
"learning_rate": 0.00012783967713437882,
|
| 645 |
+
"loss": 1.1039,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4744525547445255,
|
| 650 |
+
"grad_norm": 0.07264336315306943,
|
| 651 |
+
"learning_rate": 0.0001260869620532601,
|
| 652 |
+
"loss": 1.0703,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4796663190823775,
|
| 657 |
+
"grad_norm": 0.0716348036748029,
|
| 658 |
+
"learning_rate": 0.00012432560466535362,
|
| 659 |
+
"loss": 1.078,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4848800834202294,
|
| 664 |
+
"grad_norm": 0.07275630175563283,
|
| 665 |
+
"learning_rate": 0.00012255618848785378,
|
| 666 |
+
"loss": 1.0931,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.49009384775808135,
|
| 671 |
+
"grad_norm": 0.08738232111757031,
|
| 672 |
+
"learning_rate": 0.00012077929970773823,
|
| 673 |
+
"loss": 1.1046,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.49530761209593327,
|
| 678 |
+
"grad_norm": 0.07302314407960024,
|
| 679 |
+
"learning_rate": 0.00011899552698757078,
|
| 680 |
+
"loss": 1.0797,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5005213764337852,
|
| 685 |
+
"grad_norm": 0.07630391551541794,
|
| 686 |
+
"learning_rate": 0.00011720546127048452,
|
| 687 |
+
"loss": 1.0853,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5057351407716372,
|
| 692 |
+
"grad_norm": 0.0786638943841889,
|
| 693 |
+
"learning_rate": 0.0001154096955844091,
|
| 694 |
+
"loss": 1.0936,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5109489051094891,
|
| 699 |
+
"grad_norm": 0.0738033508924629,
|
| 700 |
+
"learning_rate": 0.00011360882484560755,
|
| 701 |
+
"loss": 1.0947,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.516162669447341,
|
| 706 |
+
"grad_norm": 0.07194256539748715,
|
| 707 |
+
"learning_rate": 0.00011180344566158739,
|
| 708 |
+
"loss": 1.0567,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5213764337851929,
|
| 713 |
+
"grad_norm": 0.07372207865811028,
|
| 714 |
+
"learning_rate": 0.0001099941561334515,
|
| 715 |
+
"loss": 1.096,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5265901981230449,
|
| 720 |
+
"grad_norm": 0.08419340992248064,
|
| 721 |
+
"learning_rate": 0.00010818155565775443,
|
| 722 |
+
"loss": 1.0906,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.5318039624608968,
|
| 727 |
+
"grad_norm": 0.07572982107794511,
|
| 728 |
+
"learning_rate": 0.00010636624472792889,
|
| 729 |
+
"loss": 1.0672,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5370177267987487,
|
| 734 |
+
"grad_norm": 0.07601878254011342,
|
| 735 |
+
"learning_rate": 0.00010454882473534961,
|
| 736 |
+
"loss": 1.0828,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5422314911366006,
|
| 741 |
+
"grad_norm": 0.07348182851139312,
|
| 742 |
+
"learning_rate": 0.0001027298977700992,
|
| 743 |
+
"loss": 1.0725,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5474452554744526,
|
| 748 |
+
"grad_norm": 0.0744277902361585,
|
| 749 |
+
"learning_rate": 0.0001009100664215028,
|
| 750 |
+
"loss": 1.0895,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5526590198123045,
|
| 755 |
+
"grad_norm": 0.07570350615821111,
|
| 756 |
+
"learning_rate": 9.908993357849721e-05,
|
| 757 |
+
"loss": 1.0873,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5578727841501564,
|
| 762 |
+
"grad_norm": 0.07422866916118566,
|
| 763 |
+
"learning_rate": 9.727010222990082e-05,
|
| 764 |
+
"loss": 1.0812,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5630865484880083,
|
| 769 |
+
"grad_norm": 0.07362006062815561,
|
| 770 |
+
"learning_rate": 9.545117526465041e-05,
|
| 771 |
+
"loss": 1.0905,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5683003128258602,
|
| 776 |
+
"grad_norm": 0.0700033281234262,
|
| 777 |
+
"learning_rate": 9.363375527207111e-05,
|
| 778 |
+
"loss": 1.0682,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5735140771637122,
|
| 783 |
+
"grad_norm": 0.07230470722040953,
|
| 784 |
+
"learning_rate": 9.181844434224558e-05,
|
| 785 |
+
"loss": 1.0873,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5787278415015641,
|
| 790 |
+
"grad_norm": 0.08088691708607157,
|
| 791 |
+
"learning_rate": 9.000584386654853e-05,
|
| 792 |
+
"loss": 1.0785,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.583941605839416,
|
| 797 |
+
"grad_norm": 0.0744369753893106,
|
| 798 |
+
"learning_rate": 8.819655433841262e-05,
|
| 799 |
+
"loss": 1.107,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5891553701772679,
|
| 804 |
+
"grad_norm": 0.07574963261021694,
|
| 805 |
+
"learning_rate": 8.639117515439248e-05,
|
| 806 |
+
"loss": 1.0854,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.59436913451512,
|
| 811 |
+
"grad_norm": 0.07470955964117636,
|
| 812 |
+
"learning_rate": 8.459030441559091e-05,
|
| 813 |
+
"loss": 1.1039,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5995828988529719,
|
| 818 |
+
"grad_norm": 0.0753809437283073,
|
| 819 |
+
"learning_rate": 8.27945387295155e-05,
|
| 820 |
+
"loss": 1.097,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6047966631908238,
|
| 825 |
+
"grad_norm": 0.07151693665161514,
|
| 826 |
+
"learning_rate": 8.100447301242923e-05,
|
| 827 |
+
"loss": 1.1062,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.6100104275286757,
|
| 832 |
+
"grad_norm": 0.07299808736310547,
|
| 833 |
+
"learning_rate": 7.92207002922618e-05,
|
| 834 |
+
"loss": 1.1024,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6152241918665277,
|
| 839 |
+
"grad_norm": 0.07463126953374469,
|
| 840 |
+
"learning_rate": 7.744381151214627e-05,
|
| 841 |
+
"loss": 1.0809,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6204379562043796,
|
| 846 |
+
"grad_norm": 0.07475965582424028,
|
| 847 |
+
"learning_rate": 7.567439533464639e-05,
|
| 848 |
+
"loss": 1.0708,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6256517205422315,
|
| 853 |
+
"grad_norm": 0.0733039750634976,
|
| 854 |
+
"learning_rate": 7.391303794673992e-05,
|
| 855 |
+
"loss": 1.1059,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6308654848800834,
|
| 860 |
+
"grad_norm": 0.08113303731253642,
|
| 861 |
+
"learning_rate": 7.216032286562122e-05,
|
| 862 |
+
"loss": 1.0891,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6360792492179353,
|
| 867 |
+
"grad_norm": 0.07598750215168182,
|
| 868 |
+
"learning_rate": 7.041683074538916e-05,
|
| 869 |
+
"loss": 1.0951,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6412930135557873,
|
| 874 |
+
"grad_norm": 0.07271525090431738,
|
| 875 |
+
"learning_rate": 6.868313918468305e-05,
|
| 876 |
+
"loss": 1.0708,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6465067778936392,
|
| 881 |
+
"grad_norm": 0.07349014640316023,
|
| 882 |
+
"learning_rate": 6.69598225353309e-05,
|
| 883 |
+
"loss": 1.0856,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6517205422314911,
|
| 888 |
+
"grad_norm": 0.07174197708446833,
|
| 889 |
+
"learning_rate": 6.524745171207339e-05,
|
| 890 |
+
"loss": 1.085,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.656934306569343,
|
| 895 |
+
"grad_norm": 0.07579799503336161,
|
| 896 |
+
"learning_rate": 6.354659400342653e-05,
|
| 897 |
+
"loss": 1.0805,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.662148070907195,
|
| 902 |
+
"grad_norm": 0.07866794639764096,
|
| 903 |
+
"learning_rate": 6.185781288374545e-05,
|
| 904 |
+
"loss": 1.0852,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6673618352450469,
|
| 909 |
+
"grad_norm": 0.07476193110508485,
|
| 910 |
+
"learning_rate": 6.0181667826551924e-05,
|
| 911 |
+
"loss": 1.0896,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6725755995828988,
|
| 916 |
+
"grad_norm": 0.0772747671129761,
|
| 917 |
+
"learning_rate": 5.851871411918743e-05,
|
| 918 |
+
"loss": 1.0792,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6777893639207507,
|
| 923 |
+
"grad_norm": 0.07566640995898331,
|
| 924 |
+
"learning_rate": 5.6869502678852835e-05,
|
| 925 |
+
"loss": 1.0471,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6830031282586028,
|
| 930 |
+
"grad_norm": 0.07164138898307278,
|
| 931 |
+
"learning_rate": 5.523457987009595e-05,
|
| 932 |
+
"loss": 1.1035,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6882168925964547,
|
| 937 |
+
"grad_norm": 0.07099725484474564,
|
| 938 |
+
"learning_rate": 5.3614487323807195e-05,
|
| 939 |
+
"loss": 1.0992,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.6934306569343066,
|
| 944 |
+
"grad_norm": 0.07751613627358377,
|
| 945 |
+
"learning_rate": 5.20097617577839e-05,
|
| 946 |
+
"loss": 1.0837,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.6986444212721585,
|
| 951 |
+
"grad_norm": 0.07882659762962813,
|
| 952 |
+
"learning_rate": 5.042093479892158e-05,
|
| 953 |
+
"loss": 1.081,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7038581856100105,
|
| 958 |
+
"grad_norm": 0.07504468684412657,
|
| 959 |
+
"learning_rate": 4.8848532807092416e-05,
|
| 960 |
+
"loss": 1.0711,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7090719499478624,
|
| 965 |
+
"grad_norm": 0.0747790732557037,
|
| 966 |
+
"learning_rate": 4.729307670076826e-05,
|
| 967 |
+
"loss": 1.077,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7142857142857143,
|
| 972 |
+
"grad_norm": 0.07214223012913477,
|
| 973 |
+
"learning_rate": 4.5755081784446306e-05,
|
| 974 |
+
"loss": 1.0894,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.7194994786235662,
|
| 979 |
+
"grad_norm": 0.07363920926565273,
|
| 980 |
+
"learning_rate": 4.4235057577934905e-05,
|
| 981 |
+
"loss": 1.0853,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7247132429614181,
|
| 986 |
+
"grad_norm": 0.0735821627304811,
|
| 987 |
+
"learning_rate": 4.273350764755548e-05,
|
| 988 |
+
"loss": 1.082,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7299270072992701,
|
| 993 |
+
"grad_norm": 0.0738898873124254,
|
| 994 |
+
"learning_rate": 4.1250929439316824e-05,
|
| 995 |
+
"loss": 1.0773,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.735140771637122,
|
| 1000 |
+
"grad_norm": 0.07402633045377013,
|
| 1001 |
+
"learning_rate": 3.978781411411705e-05,
|
| 1002 |
+
"loss": 1.1247,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7403545359749739,
|
| 1007 |
+
"grad_norm": 0.07284695841320597,
|
| 1008 |
+
"learning_rate": 3.8344646385027837e-05,
|
| 1009 |
+
"loss": 1.0754,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.7455683003128258,
|
| 1014 |
+
"grad_norm": 0.07929374122905068,
|
| 1015 |
+
"learning_rate": 3.692190435671452e-05,
|
| 1016 |
+
"loss": 1.0795,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7507820646506778,
|
| 1021 |
+
"grad_norm": 0.0734239751966692,
|
| 1022 |
+
"learning_rate": 3.5520059367045465e-05,
|
| 1023 |
+
"loss": 1.0887,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7559958289885297,
|
| 1028 |
+
"grad_norm": 0.07435223656654096,
|
| 1029 |
+
"learning_rate": 3.413957583094358e-05,
|
| 1030 |
+
"loss": 1.0999,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.7612095933263816,
|
| 1035 |
+
"grad_norm": 0.07261210019867713,
|
| 1036 |
+
"learning_rate": 3.278091108653091e-05,
|
| 1037 |
+
"loss": 1.0832,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.7664233576642335,
|
| 1042 |
+
"grad_norm": 0.07607256887516554,
|
| 1043 |
+
"learning_rate": 3.144451524361779e-05,
|
| 1044 |
+
"loss": 1.1057,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7716371220020855,
|
| 1049 |
+
"grad_norm": 0.07249789517904658,
|
| 1050 |
+
"learning_rate": 3.013083103458676e-05,
|
| 1051 |
+
"loss": 1.0786,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7768508863399375,
|
| 1056 |
+
"grad_norm": 0.07092399285995109,
|
| 1057 |
+
"learning_rate": 2.8840293667720653e-05,
|
| 1058 |
+
"loss": 1.0361,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7820646506777894,
|
| 1063 |
+
"grad_norm": 0.07310960304099395,
|
| 1064 |
+
"learning_rate": 2.7573330683022847e-05,
|
| 1065 |
+
"loss": 1.0902,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7872784150156413,
|
| 1070 |
+
"grad_norm": 0.07182455138220446,
|
| 1071 |
+
"learning_rate": 2.6330361810578395e-05,
|
| 1072 |
+
"loss": 1.0864,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7924921793534933,
|
| 1077 |
+
"grad_norm": 0.07508985054363387,
|
| 1078 |
+
"learning_rate": 2.511179883150242e-05,
|
| 1079 |
+
"loss": 1.1085,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.7977059436913452,
|
| 1084 |
+
"grad_norm": 0.07685850755584918,
|
| 1085 |
+
"learning_rate": 2.3918045441521718e-05,
|
| 1086 |
+
"loss": 1.1116,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8029197080291971,
|
| 1091 |
+
"grad_norm": 0.07574318675677963,
|
| 1092 |
+
"learning_rate": 2.274949711723502e-05,
|
| 1093 |
+
"loss": 1.0735,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.808133472367049,
|
| 1098 |
+
"grad_norm": 0.07349993942522902,
|
| 1099 |
+
"learning_rate": 2.1606540985096312e-05,
|
| 1100 |
+
"loss": 1.0892,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8133472367049009,
|
| 1105 |
+
"grad_norm": 0.07510810135075115,
|
| 1106 |
+
"learning_rate": 2.0489555693164342e-05,
|
| 1107 |
+
"loss": 1.0929,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8185610010427529,
|
| 1112 |
+
"grad_norm": 0.07465217855129215,
|
| 1113 |
+
"learning_rate": 1.9398911285660816e-05,
|
| 1114 |
+
"loss": 1.0974,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8237747653806048,
|
| 1119 |
+
"grad_norm": 0.07641853556955017,
|
| 1120 |
+
"learning_rate": 1.8334969080379115e-05,
|
| 1121 |
+
"loss": 1.0981,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8289885297184567,
|
| 1126 |
+
"grad_norm": 0.07374429336005603,
|
| 1127 |
+
"learning_rate": 1.7298081548984002e-05,
|
| 1128 |
+
"loss": 1.1147,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8342022940563086,
|
| 1133 |
+
"grad_norm": 0.07198920669236027,
|
| 1134 |
+
"learning_rate": 1.628859220024177e-05,
|
| 1135 |
+
"loss": 1.0805,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.8394160583941606,
|
| 1140 |
+
"grad_norm": 0.07357888739399879,
|
| 1141 |
+
"learning_rate": 1.5306835466219738e-05,
|
| 1142 |
+
"loss": 1.0708,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8446298227320125,
|
| 1147 |
+
"grad_norm": 0.07321249998664951,
|
| 1148 |
+
"learning_rate": 1.4353136591492933e-05,
|
| 1149 |
+
"loss": 1.1037,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8498435870698644,
|
| 1154 |
+
"grad_norm": 0.07782709844758377,
|
| 1155 |
+
"learning_rate": 1.3427811525394319e-05,
|
| 1156 |
+
"loss": 1.0904,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.8550573514077163,
|
| 1161 |
+
"grad_norm": 0.07455259445795498,
|
| 1162 |
+
"learning_rate": 1.253116681734432e-05,
|
| 1163 |
+
"loss": 1.0768,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8602711157455682,
|
| 1168 |
+
"grad_norm": 0.07401992962356452,
|
| 1169 |
+
"learning_rate": 1.1663499515294762e-05,
|
| 1170 |
+
"loss": 1.0653,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8654848800834203,
|
| 1175 |
+
"grad_norm": 0.07220555107508732,
|
| 1176 |
+
"learning_rate": 1.0825097067320267e-05,
|
| 1177 |
+
"loss": 1.0729,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8706986444212722,
|
| 1182 |
+
"grad_norm": 0.07544891513968513,
|
| 1183 |
+
"learning_rate": 1.0016237226389947e-05,
|
| 1184 |
+
"loss": 1.0855,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.8759124087591241,
|
| 1189 |
+
"grad_norm": 0.07485172694158737,
|
| 1190 |
+
"learning_rate": 9.237187958351279e-06,
|
| 1191 |
+
"loss": 1.09,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.881126173096976,
|
| 1196 |
+
"grad_norm": 0.07275386970426083,
|
| 1197 |
+
"learning_rate": 8.488207353155986e-06,
|
| 1198 |
+
"loss": 1.0896,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.886339937434828,
|
| 1203 |
+
"grad_norm": 0.07454861219445595,
|
| 1204 |
+
"learning_rate": 7.769543539357904e-06,
|
| 1205 |
+
"loss": 1.0953,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.8915537017726799,
|
| 1210 |
+
"grad_norm": 0.07164206957934503,
|
| 1211 |
+
"learning_rate": 7.081434601910864e-06,
|
| 1212 |
+
"loss": 1.0944,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8967674661105318,
|
| 1217 |
+
"grad_norm": 0.07286862134246115,
|
| 1218 |
+
"learning_rate": 6.424108503293924e-06,
|
| 1219 |
+
"loss": 1.1096,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9019812304483837,
|
| 1224 |
+
"grad_norm": 0.07369229576160694,
|
| 1225 |
+
"learning_rate": 5.797783007990076e-06,
|
| 1226 |
+
"loss": 1.0951,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.9071949947862357,
|
| 1231 |
+
"grad_norm": 0.07460818380676967,
|
| 1232 |
+
"learning_rate": 5.202665610343338e-06,
|
| 1233 |
+
"loss": 1.0883,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.9124087591240876,
|
| 1238 |
+
"grad_norm": 0.07357667175850421,
|
| 1239 |
+
"learning_rate": 4.6389534658184075e-06,
|
| 1240 |
+
"loss": 1.0856,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.9176225234619395,
|
| 1245 |
+
"grad_norm": 0.07355488764311813,
|
| 1246 |
+
"learning_rate": 4.106833325685222e-06,
|
| 1247 |
+
"loss": 1.0992,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9228362877997914,
|
| 1252 |
+
"grad_norm": 0.07560170907908333,
|
| 1253 |
+
"learning_rate": 3.606481475150536e-06,
|
| 1254 |
+
"loss": 1.0753,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.9280500521376434,
|
| 1259 |
+
"grad_norm": 0.07352332452876986,
|
| 1260 |
+
"learning_rate": 3.138063674956648e-06,
|
| 1261 |
+
"loss": 1.0606,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9332638164754953,
|
| 1266 |
+
"grad_norm": 0.07595937879635985,
|
| 1267 |
+
"learning_rate": 2.701735106466796e-06,
|
| 1268 |
+
"loss": 1.0931,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9384775808133472,
|
| 1273 |
+
"grad_norm": 0.07046192596564868,
|
| 1274 |
+
"learning_rate": 2.2976403202553787e-06,
|
| 1275 |
+
"loss": 1.0931,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9436913451511991,
|
| 1280 |
+
"grad_norm": 0.07358829064156538,
|
| 1281 |
+
"learning_rate": 1.925913188220052e-06,
|
| 1282 |
+
"loss": 1.0806,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.948905109489051,
|
| 1287 |
+
"grad_norm": 0.07595473779694568,
|
| 1288 |
+
"learning_rate": 1.586676859231473e-06,
|
| 1289 |
+
"loss": 1.0816,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.954118873826903,
|
| 1294 |
+
"grad_norm": 0.07568884330826342,
|
| 1295 |
+
"learning_rate": 1.2800437183355885e-06,
|
| 1296 |
+
"loss": 1.0809,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.959332638164755,
|
| 1301 |
+
"grad_norm": 0.07584100660095472,
|
| 1302 |
+
"learning_rate": 1.0061153495217412e-06,
|
| 1303 |
+
"loss": 1.0634,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9645464025026069,
|
| 1308 |
+
"grad_norm": 0.07304112910406586,
|
| 1309 |
+
"learning_rate": 7.64982502069056e-07,
|
| 1310 |
+
"loss": 1.0592,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.9697601668404588,
|
| 1315 |
+
"grad_norm": 0.07324941671378657,
|
| 1316 |
+
"learning_rate": 5.567250604822882e-07,
|
| 1317 |
+
"loss": 1.0764,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9749739311783108,
|
| 1322 |
+
"grad_norm": 0.07540315428759628,
|
| 1323 |
+
"learning_rate": 3.814120180270164e-07,
|
| 1324 |
+
"loss": 1.0862,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.9801876955161627,
|
| 1329 |
+
"grad_norm": 0.07167856975306086,
|
| 1330 |
+
"learning_rate": 2.3910145387299185e-07,
|
| 1331 |
+
"loss": 1.0798,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.9854014598540146,
|
| 1336 |
+
"grad_norm": 0.07399712911952666,
|
| 1337 |
+
"learning_rate": 1.298405138531078e-07,
|
| 1338 |
+
"loss": 1.0819,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.9906152241918665,
|
| 1343 |
+
"grad_norm": 0.07550044600887201,
|
| 1344 |
+
"learning_rate": 5.366539484464861e-08,
|
| 1345 |
+
"loss": 1.088,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.9958289885297185,
|
| 1350 |
+
"grad_norm": 0.07533429394464258,
|
| 1351 |
+
"learning_rate": 1.0601332777604444e-08,
|
| 1352 |
+
"loss": 1.071,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.0,
|
| 1357 |
+
"eval_loss": 1.1101280450820923,
|
| 1358 |
+
"eval_runtime": 2.0768,
|
| 1359 |
+
"eval_samples_per_second": 3.371,
|
| 1360 |
+
"eval_steps_per_second": 0.963,
|
| 1361 |
+
"step": 959
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.0,
|
| 1365 |
+
"step": 959,
|
| 1366 |
+
"total_flos": 1.2060020938113024e+16,
|
| 1367 |
+
"train_loss": 1.1071907986689657,
|
| 1368 |
+
"train_runtime": 19793.4474,
|
| 1369 |
+
"train_samples_per_second": 3.101,
|
| 1370 |
+
"train_steps_per_second": 0.048
|
| 1371 |
+
}
|
| 1372 |
+
],
|
| 1373 |
+
"logging_steps": 5,
|
| 1374 |
+
"max_steps": 959,
|
| 1375 |
+
"num_input_tokens_seen": 0,
|
| 1376 |
+
"num_train_epochs": 1,
|
| 1377 |
+
"save_steps": 100,
|
| 1378 |
+
"stateful_callbacks": {
|
| 1379 |
+
"TrainerControl": {
|
| 1380 |
+
"args": {
|
| 1381 |
+
"should_epoch_stop": false,
|
| 1382 |
+
"should_evaluate": false,
|
| 1383 |
+
"should_log": false,
|
| 1384 |
+
"should_save": true,
|
| 1385 |
+
"should_training_stop": true
|
| 1386 |
+
},
|
| 1387 |
+
"attributes": {}
|
| 1388 |
+
}
|
| 1389 |
+
},
|
| 1390 |
+
"total_flos": 1.2060020938113024e+16,
|
| 1391 |
+
"train_batch_size": 4,
|
| 1392 |
+
"trial_name": null,
|
| 1393 |
+
"trial_params": null
|
| 1394 |
+
}
|