Model save
Browse files- README.md +67 -0
- trainer_state.json +2862 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: openbmb/MiniCPM-V-2_6
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: MiniCPM-V-2_6_lora_20240917_011727
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# MiniCPM-V-2_6_lora_20240917_011727
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [openbmb/MiniCPM-V-2_6](https://huggingface.co/openbmb/MiniCPM-V-2_6) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 1.3224
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 5e-05
|
| 38 |
+
- train_batch_size: 2
|
| 39 |
+
- eval_batch_size: 2
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 4
|
| 43 |
+
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- total_eval_batch_size: 8
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine_with_restarts
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.05
|
| 49 |
+
- training_steps: 1000
|
| 50 |
+
- mixed_precision_training: Native AMP
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 56 |
+
| 1.3412 | 0.0349 | 100 | 1.3380 |
|
| 57 |
+
| 1.3477 | 0.0698 | 200 | 1.3300 |
|
| 58 |
+
| 1.3057 | 0.1047 | 300 | 1.3255 |
|
| 59 |
+
| 1.3116 | 0.1396 | 400 | 1.3224 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- PEFT 0.12.0
|
| 65 |
+
- Transformers 4.40.0
|
| 66 |
+
- Pytorch 2.1.2+cu121
|
| 67 |
+
- Tokenizers 0.19.1
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2862 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.322394609451294,
|
| 3 |
+
"best_model_checkpoint": "output/output__lora/checkpoint-400",
|
| 4 |
+
"epoch": 0.139640425903299,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00034910106475824753,
|
| 13 |
+
"grad_norm": 2.6783504486083984,
|
| 14 |
+
"learning_rate": 0.0,
|
| 15 |
+
"loss": 1.5271,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0006982021295164951,
|
| 20 |
+
"grad_norm": 1.3333820104599,
|
| 21 |
+
"learning_rate": 8.859191006777897e-06,
|
| 22 |
+
"loss": 1.3963,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0010473031942747426,
|
| 27 |
+
"grad_norm": 1.2807133197784424,
|
| 28 |
+
"learning_rate": 1.4041485532469073e-05,
|
| 29 |
+
"loss": 1.4192,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0013964042590329901,
|
| 34 |
+
"grad_norm": 1.1956514120101929,
|
| 35 |
+
"learning_rate": 1.7718382013555794e-05,
|
| 36 |
+
"loss": 1.5083,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0017455053237912376,
|
| 41 |
+
"grad_norm": 1.2733005285263062,
|
| 42 |
+
"learning_rate": 2.0570404496611053e-05,
|
| 43 |
+
"loss": 1.4963,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0020946063885494853,
|
| 48 |
+
"grad_norm": 0.8666600584983826,
|
| 49 |
+
"learning_rate": 2.2900676539246968e-05,
|
| 50 |
+
"loss": 1.5552,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0024437074533077328,
|
| 55 |
+
"grad_norm": 0.7445533275604248,
|
| 56 |
+
"learning_rate": 2.4870893478326387e-05,
|
| 57 |
+
"loss": 1.2858,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0027928085180659802,
|
| 62 |
+
"grad_norm": 0.8400186896324158,
|
| 63 |
+
"learning_rate": 2.6577573020333684e-05,
|
| 64 |
+
"loss": 1.3413,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0031419095828242277,
|
| 69 |
+
"grad_norm": 0.8454774618148804,
|
| 70 |
+
"learning_rate": 2.8082971064938146e-05,
|
| 71 |
+
"loss": 1.467,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.003491010647582475,
|
| 76 |
+
"grad_norm": 0.8853550553321838,
|
| 77 |
+
"learning_rate": 2.9429595503388953e-05,
|
| 78 |
+
"loss": 1.4477,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0038401117123407227,
|
| 83 |
+
"grad_norm": 1.4953877925872803,
|
| 84 |
+
"learning_rate": 3.064776548439465e-05,
|
| 85 |
+
"loss": 1.4012,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0041892127770989706,
|
| 90 |
+
"grad_norm": 0.8356307148933411,
|
| 91 |
+
"learning_rate": 3.1759867546024865e-05,
|
| 92 |
+
"loss": 1.3855,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.004538313841857218,
|
| 97 |
+
"grad_norm": 0.7591987252235413,
|
| 98 |
+
"learning_rate": 3.2782902272079295e-05,
|
| 99 |
+
"loss": 1.3561,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0048874149066154655,
|
| 104 |
+
"grad_norm": 0.9811077117919922,
|
| 105 |
+
"learning_rate": 3.373008448510428e-05,
|
| 106 |
+
"loss": 1.3175,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.005236515971373713,
|
| 111 |
+
"grad_norm": 0.8403587341308594,
|
| 112 |
+
"learning_rate": 3.4611890029080124e-05,
|
| 113 |
+
"loss": 1.341,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0055856170361319605,
|
| 118 |
+
"grad_norm": 0.750234067440033,
|
| 119 |
+
"learning_rate": 3.543676402711159e-05,
|
| 120 |
+
"loss": 1.4247,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.005934718100890208,
|
| 125 |
+
"grad_norm": 0.7567417621612549,
|
| 126 |
+
"learning_rate": 3.621161404374383e-05,
|
| 127 |
+
"loss": 1.416,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.006283819165648455,
|
| 132 |
+
"grad_norm": 0.7126427292823792,
|
| 133 |
+
"learning_rate": 3.694216207171603e-05,
|
| 134 |
+
"loss": 1.4426,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.006632920230406703,
|
| 139 |
+
"grad_norm": 0.7808831930160522,
|
| 140 |
+
"learning_rate": 3.76332012245438e-05,
|
| 141 |
+
"loss": 1.4287,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.00698202129516495,
|
| 146 |
+
"grad_norm": 0.6165328025817871,
|
| 147 |
+
"learning_rate": 3.8288786510166846e-05,
|
| 148 |
+
"loss": 1.3391,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.007331122359923198,
|
| 153 |
+
"grad_norm": 0.7212307453155518,
|
| 154 |
+
"learning_rate": 3.8912379010795455e-05,
|
| 155 |
+
"loss": 1.3375,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.007680223424681445,
|
| 160 |
+
"grad_norm": 0.6797880530357361,
|
| 161 |
+
"learning_rate": 3.9506956491172545e-05,
|
| 162 |
+
"loss": 1.2713,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.008029324489439693,
|
| 167 |
+
"grad_norm": 0.7757507562637329,
|
| 168 |
+
"learning_rate": 4.007509939970292e-05,
|
| 169 |
+
"loss": 1.3599,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.008378425554197941,
|
| 174 |
+
"grad_norm": 0.539090096950531,
|
| 175 |
+
"learning_rate": 4.061905855280276e-05,
|
| 176 |
+
"loss": 1.5154,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.008727526618956188,
|
| 181 |
+
"grad_norm": 0.652180552482605,
|
| 182 |
+
"learning_rate": 4.1140808993222106e-05,
|
| 183 |
+
"loss": 1.3438,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.009076627683714436,
|
| 188 |
+
"grad_norm": 0.7319611310958862,
|
| 189 |
+
"learning_rate": 4.164209327885719e-05,
|
| 190 |
+
"loss": 1.5033,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.009425728748472683,
|
| 195 |
+
"grad_norm": 0.702570378780365,
|
| 196 |
+
"learning_rate": 4.2124456597407214e-05,
|
| 197 |
+
"loss": 1.2238,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.009774829813230931,
|
| 202 |
+
"grad_norm": 0.6835883855819702,
|
| 203 |
+
"learning_rate": 4.258927549188218e-05,
|
| 204 |
+
"loss": 1.3648,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.010123930877989178,
|
| 209 |
+
"grad_norm": 0.6773353219032288,
|
| 210 |
+
"learning_rate": 4.303778154313212e-05,
|
| 211 |
+
"loss": 1.3074,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.010473031942747426,
|
| 216 |
+
"grad_norm": 0.6387542486190796,
|
| 217 |
+
"learning_rate": 4.347108103585803e-05,
|
| 218 |
+
"loss": 1.2265,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.010822133007505673,
|
| 223 |
+
"grad_norm": 0.6249099969863892,
|
| 224 |
+
"learning_rate": 4.389017139879164e-05,
|
| 225 |
+
"loss": 1.3321,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.011171234072263921,
|
| 230 |
+
"grad_norm": 0.7121676802635193,
|
| 231 |
+
"learning_rate": 4.429595503388948e-05,
|
| 232 |
+
"loss": 1.3729,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.011520335137022168,
|
| 237 |
+
"grad_norm": 0.7367205619812012,
|
| 238 |
+
"learning_rate": 4.468925101686371e-05,
|
| 239 |
+
"loss": 1.3937,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.011869436201780416,
|
| 244 |
+
"grad_norm": 0.6183043718338013,
|
| 245 |
+
"learning_rate": 4.507080505052173e-05,
|
| 246 |
+
"loss": 1.4321,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.012218537266538662,
|
| 251 |
+
"grad_norm": 1.1439142227172852,
|
| 252 |
+
"learning_rate": 4.544129797493744e-05,
|
| 253 |
+
"loss": 1.3515,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.01256763833129691,
|
| 258 |
+
"grad_norm": 0.7980801463127136,
|
| 259 |
+
"learning_rate": 4.5801353078493936e-05,
|
| 260 |
+
"loss": 1.3929,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.012916739396055157,
|
| 265 |
+
"grad_norm": 0.8890343904495239,
|
| 266 |
+
"learning_rate": 4.615154240700883e-05,
|
| 267 |
+
"loss": 1.2895,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.013265840460813406,
|
| 272 |
+
"grad_norm": 0.7107703685760498,
|
| 273 |
+
"learning_rate": 4.6492392231321696e-05,
|
| 274 |
+
"loss": 1.3054,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.013614941525571652,
|
| 279 |
+
"grad_norm": 0.605403482913971,
|
| 280 |
+
"learning_rate": 4.682438780454837e-05,
|
| 281 |
+
"loss": 1.3817,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.0139640425903299,
|
| 286 |
+
"grad_norm": 0.6489142775535583,
|
| 287 |
+
"learning_rate": 4.714797751694474e-05,
|
| 288 |
+
"loss": 1.4109,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.014313143655088147,
|
| 293 |
+
"grad_norm": 0.5896831750869751,
|
| 294 |
+
"learning_rate": 4.7463576537657414e-05,
|
| 295 |
+
"loss": 1.3383,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.014662244719846396,
|
| 300 |
+
"grad_norm": 0.8319935202598572,
|
| 301 |
+
"learning_rate": 4.777157001757336e-05,
|
| 302 |
+
"loss": 1.4239,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.015011345784604642,
|
| 307 |
+
"grad_norm": 0.6128418445587158,
|
| 308 |
+
"learning_rate": 4.8072315915252694e-05,
|
| 309 |
+
"loss": 1.3541,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.01536044684936289,
|
| 314 |
+
"grad_norm": 0.6820589900016785,
|
| 315 |
+
"learning_rate": 4.8366147497950435e-05,
|
| 316 |
+
"loss": 1.2663,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.015709547914121137,
|
| 321 |
+
"grad_norm": 0.8375743627548218,
|
| 322 |
+
"learning_rate": 4.8653375561549195e-05,
|
| 323 |
+
"loss": 1.3803,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.016058648978879386,
|
| 328 |
+
"grad_norm": 0.6585806608200073,
|
| 329 |
+
"learning_rate": 4.8934290406480814e-05,
|
| 330 |
+
"loss": 1.3143,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.016407750043637634,
|
| 335 |
+
"grad_norm": 0.7528412342071533,
|
| 336 |
+
"learning_rate": 4.920916360113129e-05,
|
| 337 |
+
"loss": 1.293,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.016756851108395882,
|
| 342 |
+
"grad_norm": 0.6918306946754456,
|
| 343 |
+
"learning_rate": 4.947824955958066e-05,
|
| 344 |
+
"loss": 1.4991,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.017105952173154127,
|
| 349 |
+
"grad_norm": 0.6764557361602783,
|
| 350 |
+
"learning_rate": 4.9741786956652774e-05,
|
| 351 |
+
"loss": 1.2755,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.017455053237912375,
|
| 356 |
+
"grad_norm": 0.6525936722755432,
|
| 357 |
+
"learning_rate": 5e-05,
|
| 358 |
+
"loss": 1.3897,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.017804154302670624,
|
| 363 |
+
"grad_norm": 0.627804160118103,
|
| 364 |
+
"learning_rate": 5e-05,
|
| 365 |
+
"loss": 1.3027,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.018153255367428872,
|
| 370 |
+
"grad_norm": 0.8060218095779419,
|
| 371 |
+
"learning_rate": 5e-05,
|
| 372 |
+
"loss": 1.3477,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.018502356432187117,
|
| 377 |
+
"grad_norm": 0.6655098795890808,
|
| 378 |
+
"learning_rate": 5e-05,
|
| 379 |
+
"loss": 1.3631,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.018851457496945365,
|
| 384 |
+
"grad_norm": 0.7165637016296387,
|
| 385 |
+
"learning_rate": 5e-05,
|
| 386 |
+
"loss": 1.347,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.019200558561703614,
|
| 391 |
+
"grad_norm": 0.6562020778656006,
|
| 392 |
+
"learning_rate": 5e-05,
|
| 393 |
+
"loss": 1.3535,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.019549659626461862,
|
| 398 |
+
"grad_norm": 0.7588657736778259,
|
| 399 |
+
"learning_rate": 5e-05,
|
| 400 |
+
"loss": 1.3291,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.019898760691220107,
|
| 405 |
+
"grad_norm": 0.6295105814933777,
|
| 406 |
+
"learning_rate": 5e-05,
|
| 407 |
+
"loss": 1.3542,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.020247861755978355,
|
| 412 |
+
"grad_norm": 1.339097023010254,
|
| 413 |
+
"learning_rate": 5e-05,
|
| 414 |
+
"loss": 1.3649,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.020596962820736604,
|
| 419 |
+
"grad_norm": 0.6976660490036011,
|
| 420 |
+
"learning_rate": 5e-05,
|
| 421 |
+
"loss": 1.2852,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.020946063885494852,
|
| 426 |
+
"grad_norm": 0.7590420246124268,
|
| 427 |
+
"learning_rate": 5e-05,
|
| 428 |
+
"loss": 1.354,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.021295164950253097,
|
| 433 |
+
"grad_norm": 0.6279817819595337,
|
| 434 |
+
"learning_rate": 5e-05,
|
| 435 |
+
"loss": 1.2537,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.021644266015011345,
|
| 440 |
+
"grad_norm": 0.6099221110343933,
|
| 441 |
+
"learning_rate": 5e-05,
|
| 442 |
+
"loss": 1.2423,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.021993367079769593,
|
| 447 |
+
"grad_norm": 0.6252647638320923,
|
| 448 |
+
"learning_rate": 5e-05,
|
| 449 |
+
"loss": 1.3667,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.022342468144527842,
|
| 454 |
+
"grad_norm": 0.8939846158027649,
|
| 455 |
+
"learning_rate": 5e-05,
|
| 456 |
+
"loss": 1.2889,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.022691569209286087,
|
| 461 |
+
"grad_norm": 0.85840904712677,
|
| 462 |
+
"learning_rate": 5e-05,
|
| 463 |
+
"loss": 1.3747,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.023040670274044335,
|
| 468 |
+
"grad_norm": 0.8478113412857056,
|
| 469 |
+
"learning_rate": 5e-05,
|
| 470 |
+
"loss": 1.3417,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.023389771338802583,
|
| 475 |
+
"grad_norm": 0.6869573593139648,
|
| 476 |
+
"learning_rate": 5e-05,
|
| 477 |
+
"loss": 1.4033,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.02373887240356083,
|
| 482 |
+
"grad_norm": 0.6566379070281982,
|
| 483 |
+
"learning_rate": 5e-05,
|
| 484 |
+
"loss": 1.3617,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.02408797346831908,
|
| 489 |
+
"grad_norm": 0.6871697306632996,
|
| 490 |
+
"learning_rate": 5e-05,
|
| 491 |
+
"loss": 1.2932,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.024437074533077325,
|
| 496 |
+
"grad_norm": 0.7102701663970947,
|
| 497 |
+
"learning_rate": 5e-05,
|
| 498 |
+
"loss": 1.4062,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.024786175597835573,
|
| 503 |
+
"grad_norm": 0.8392966985702515,
|
| 504 |
+
"learning_rate": 5e-05,
|
| 505 |
+
"loss": 1.1992,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.02513527666259382,
|
| 510 |
+
"grad_norm": 0.670971155166626,
|
| 511 |
+
"learning_rate": 5e-05,
|
| 512 |
+
"loss": 1.4131,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.02548437772735207,
|
| 517 |
+
"grad_norm": 0.7271628975868225,
|
| 518 |
+
"learning_rate": 5e-05,
|
| 519 |
+
"loss": 1.2928,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.025833478792110315,
|
| 524 |
+
"grad_norm": 0.7184221744537354,
|
| 525 |
+
"learning_rate": 5e-05,
|
| 526 |
+
"loss": 1.2239,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.026182579856868563,
|
| 531 |
+
"grad_norm": 0.5685485005378723,
|
| 532 |
+
"learning_rate": 5e-05,
|
| 533 |
+
"loss": 1.2692,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.02653168092162681,
|
| 538 |
+
"grad_norm": 0.5677881836891174,
|
| 539 |
+
"learning_rate": 5e-05,
|
| 540 |
+
"loss": 1.2951,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.02688078198638506,
|
| 545 |
+
"grad_norm": 0.6896436810493469,
|
| 546 |
+
"learning_rate": 5e-05,
|
| 547 |
+
"loss": 1.3297,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.027229883051143305,
|
| 552 |
+
"grad_norm": 0.6284964084625244,
|
| 553 |
+
"learning_rate": 5e-05,
|
| 554 |
+
"loss": 1.2402,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.027578984115901553,
|
| 559 |
+
"grad_norm": 0.618015468120575,
|
| 560 |
+
"learning_rate": 5e-05,
|
| 561 |
+
"loss": 1.2999,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.0279280851806598,
|
| 566 |
+
"grad_norm": 0.7585094571113586,
|
| 567 |
+
"learning_rate": 5e-05,
|
| 568 |
+
"loss": 1.3378,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.02827718624541805,
|
| 573 |
+
"grad_norm": 0.6674929857254028,
|
| 574 |
+
"learning_rate": 5e-05,
|
| 575 |
+
"loss": 1.3585,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.028626287310176295,
|
| 580 |
+
"grad_norm": 0.583121120929718,
|
| 581 |
+
"learning_rate": 5e-05,
|
| 582 |
+
"loss": 1.3236,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.028975388374934543,
|
| 587 |
+
"grad_norm": 0.661668062210083,
|
| 588 |
+
"learning_rate": 5e-05,
|
| 589 |
+
"loss": 1.3264,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.02932448943969279,
|
| 594 |
+
"grad_norm": 0.8168457746505737,
|
| 595 |
+
"learning_rate": 5e-05,
|
| 596 |
+
"loss": 1.3132,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.02967359050445104,
|
| 601 |
+
"grad_norm": 0.6123843193054199,
|
| 602 |
+
"learning_rate": 5e-05,
|
| 603 |
+
"loss": 1.3224,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.030022691569209285,
|
| 608 |
+
"grad_norm": 0.7081793546676636,
|
| 609 |
+
"learning_rate": 5e-05,
|
| 610 |
+
"loss": 1.3641,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.030371792633967533,
|
| 615 |
+
"grad_norm": 0.7772612571716309,
|
| 616 |
+
"learning_rate": 5e-05,
|
| 617 |
+
"loss": 1.3634,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.03072089369872578,
|
| 622 |
+
"grad_norm": 0.603370726108551,
|
| 623 |
+
"learning_rate": 5e-05,
|
| 624 |
+
"loss": 1.4486,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.03106999476348403,
|
| 629 |
+
"grad_norm": 0.6567598581314087,
|
| 630 |
+
"learning_rate": 5e-05,
|
| 631 |
+
"loss": 1.4228,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.031419095828242274,
|
| 636 |
+
"grad_norm": 0.6245101690292358,
|
| 637 |
+
"learning_rate": 5e-05,
|
| 638 |
+
"loss": 1.2928,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.031768196893000526,
|
| 643 |
+
"grad_norm": 0.7198782563209534,
|
| 644 |
+
"learning_rate": 5e-05,
|
| 645 |
+
"loss": 1.3304,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.03211729795775877,
|
| 650 |
+
"grad_norm": 0.526452898979187,
|
| 651 |
+
"learning_rate": 5e-05,
|
| 652 |
+
"loss": 1.3418,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.032466399022517016,
|
| 657 |
+
"grad_norm": 0.7534317374229431,
|
| 658 |
+
"learning_rate": 5e-05,
|
| 659 |
+
"loss": 1.333,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.03281550008727527,
|
| 664 |
+
"grad_norm": 0.5721869468688965,
|
| 665 |
+
"learning_rate": 5e-05,
|
| 666 |
+
"loss": 1.1849,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.03316460115203351,
|
| 671 |
+
"grad_norm": 0.6943261027336121,
|
| 672 |
+
"learning_rate": 5e-05,
|
| 673 |
+
"loss": 1.3263,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.033513702216791764,
|
| 678 |
+
"grad_norm": 0.5904171466827393,
|
| 679 |
+
"learning_rate": 5e-05,
|
| 680 |
+
"loss": 1.3103,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.03386280328155001,
|
| 685 |
+
"grad_norm": 0.7743117809295654,
|
| 686 |
+
"learning_rate": 5e-05,
|
| 687 |
+
"loss": 1.3633,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.034211904346308254,
|
| 692 |
+
"grad_norm": 1.298839807510376,
|
| 693 |
+
"learning_rate": 5e-05,
|
| 694 |
+
"loss": 1.335,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.034561005411066506,
|
| 699 |
+
"grad_norm": 0.7134571671485901,
|
| 700 |
+
"learning_rate": 5e-05,
|
| 701 |
+
"loss": 1.4154,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.03491010647582475,
|
| 706 |
+
"grad_norm": 0.6801385879516602,
|
| 707 |
+
"learning_rate": 5e-05,
|
| 708 |
+
"loss": 1.3412,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.03491010647582475,
|
| 713 |
+
"eval_loss": 1.337953805923462,
|
| 714 |
+
"eval_runtime": 3305.6905,
|
| 715 |
+
"eval_samples_per_second": 6.932,
|
| 716 |
+
"eval_steps_per_second": 0.867,
|
| 717 |
+
"step": 100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.035259207540582996,
|
| 721 |
+
"grad_norm": 1.0192288160324097,
|
| 722 |
+
"learning_rate": 5e-05,
|
| 723 |
+
"loss": 1.2821,
|
| 724 |
+
"step": 101
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.03560830860534125,
|
| 728 |
+
"grad_norm": 0.6322550773620605,
|
| 729 |
+
"learning_rate": 5e-05,
|
| 730 |
+
"loss": 1.3561,
|
| 731 |
+
"step": 102
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.03595740967009949,
|
| 735 |
+
"grad_norm": 0.6499407291412354,
|
| 736 |
+
"learning_rate": 5e-05,
|
| 737 |
+
"loss": 1.3164,
|
| 738 |
+
"step": 103
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.036306510734857744,
|
| 742 |
+
"grad_norm": 0.7576645612716675,
|
| 743 |
+
"learning_rate": 5e-05,
|
| 744 |
+
"loss": 1.2924,
|
| 745 |
+
"step": 104
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.03665561179961599,
|
| 749 |
+
"grad_norm": 0.6215568780899048,
|
| 750 |
+
"learning_rate": 5e-05,
|
| 751 |
+
"loss": 1.2551,
|
| 752 |
+
"step": 105
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.037004712864374234,
|
| 756 |
+
"grad_norm": 0.6197790503501892,
|
| 757 |
+
"learning_rate": 5e-05,
|
| 758 |
+
"loss": 1.317,
|
| 759 |
+
"step": 106
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.037353813929132486,
|
| 763 |
+
"grad_norm": 0.677772045135498,
|
| 764 |
+
"learning_rate": 5e-05,
|
| 765 |
+
"loss": 1.428,
|
| 766 |
+
"step": 107
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.03770291499389073,
|
| 770 |
+
"grad_norm": 0.6386198401451111,
|
| 771 |
+
"learning_rate": 5e-05,
|
| 772 |
+
"loss": 1.4206,
|
| 773 |
+
"step": 108
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.038052016058648976,
|
| 777 |
+
"grad_norm": 1.113053798675537,
|
| 778 |
+
"learning_rate": 5e-05,
|
| 779 |
+
"loss": 1.3992,
|
| 780 |
+
"step": 109
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.03840111712340723,
|
| 784 |
+
"grad_norm": 0.668409526348114,
|
| 785 |
+
"learning_rate": 5e-05,
|
| 786 |
+
"loss": 1.3358,
|
| 787 |
+
"step": 110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.03875021818816547,
|
| 791 |
+
"grad_norm": 0.6381022930145264,
|
| 792 |
+
"learning_rate": 5e-05,
|
| 793 |
+
"loss": 1.245,
|
| 794 |
+
"step": 111
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.039099319252923724,
|
| 798 |
+
"grad_norm": 0.7082274556159973,
|
| 799 |
+
"learning_rate": 5e-05,
|
| 800 |
+
"loss": 1.3107,
|
| 801 |
+
"step": 112
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.03944842031768197,
|
| 805 |
+
"grad_norm": 0.6497403979301453,
|
| 806 |
+
"learning_rate": 5e-05,
|
| 807 |
+
"loss": 1.3174,
|
| 808 |
+
"step": 113
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.039797521382440214,
|
| 812 |
+
"grad_norm": 0.7390655279159546,
|
| 813 |
+
"learning_rate": 5e-05,
|
| 814 |
+
"loss": 1.2791,
|
| 815 |
+
"step": 114
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.040146622447198466,
|
| 819 |
+
"grad_norm": 0.6828505992889404,
|
| 820 |
+
"learning_rate": 5e-05,
|
| 821 |
+
"loss": 1.3903,
|
| 822 |
+
"step": 115
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.04049572351195671,
|
| 826 |
+
"grad_norm": 0.6913119554519653,
|
| 827 |
+
"learning_rate": 5e-05,
|
| 828 |
+
"loss": 1.3147,
|
| 829 |
+
"step": 116
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.04084482457671496,
|
| 833 |
+
"grad_norm": 0.6394439339637756,
|
| 834 |
+
"learning_rate": 5e-05,
|
| 835 |
+
"loss": 1.3308,
|
| 836 |
+
"step": 117
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.04119392564147321,
|
| 840 |
+
"grad_norm": 0.6368663907051086,
|
| 841 |
+
"learning_rate": 5e-05,
|
| 842 |
+
"loss": 1.3021,
|
| 843 |
+
"step": 118
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.04154302670623145,
|
| 847 |
+
"grad_norm": 0.625417947769165,
|
| 848 |
+
"learning_rate": 5e-05,
|
| 849 |
+
"loss": 1.4122,
|
| 850 |
+
"step": 119
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.041892127770989704,
|
| 854 |
+
"grad_norm": 0.5640509724617004,
|
| 855 |
+
"learning_rate": 5e-05,
|
| 856 |
+
"loss": 1.3216,
|
| 857 |
+
"step": 120
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.04224122883574795,
|
| 861 |
+
"grad_norm": 0.6355682611465454,
|
| 862 |
+
"learning_rate": 5e-05,
|
| 863 |
+
"loss": 1.2522,
|
| 864 |
+
"step": 121
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.042590329900506194,
|
| 868 |
+
"grad_norm": 2.130183696746826,
|
| 869 |
+
"learning_rate": 5e-05,
|
| 870 |
+
"loss": 1.398,
|
| 871 |
+
"step": 122
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.042939430965264445,
|
| 875 |
+
"grad_norm": 0.7858290672302246,
|
| 876 |
+
"learning_rate": 5e-05,
|
| 877 |
+
"loss": 1.3543,
|
| 878 |
+
"step": 123
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.04328853203002269,
|
| 882 |
+
"grad_norm": 0.6912608742713928,
|
| 883 |
+
"learning_rate": 5e-05,
|
| 884 |
+
"loss": 1.3338,
|
| 885 |
+
"step": 124
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.04363763309478094,
|
| 889 |
+
"grad_norm": 0.6326834559440613,
|
| 890 |
+
"learning_rate": 5e-05,
|
| 891 |
+
"loss": 1.2968,
|
| 892 |
+
"step": 125
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.04398673415953919,
|
| 896 |
+
"grad_norm": 0.6076151728630066,
|
| 897 |
+
"learning_rate": 5e-05,
|
| 898 |
+
"loss": 1.2705,
|
| 899 |
+
"step": 126
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.04433583522429743,
|
| 903 |
+
"grad_norm": 0.767652153968811,
|
| 904 |
+
"learning_rate": 5e-05,
|
| 905 |
+
"loss": 1.3601,
|
| 906 |
+
"step": 127
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.044684936289055684,
|
| 910 |
+
"grad_norm": 0.621769905090332,
|
| 911 |
+
"learning_rate": 5e-05,
|
| 912 |
+
"loss": 1.2834,
|
| 913 |
+
"step": 128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.04503403735381393,
|
| 917 |
+
"grad_norm": 0.6216384768486023,
|
| 918 |
+
"learning_rate": 5e-05,
|
| 919 |
+
"loss": 1.3322,
|
| 920 |
+
"step": 129
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.04538313841857217,
|
| 924 |
+
"grad_norm": 0.626325249671936,
|
| 925 |
+
"learning_rate": 5e-05,
|
| 926 |
+
"loss": 1.4601,
|
| 927 |
+
"step": 130
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.045732239483330425,
|
| 931 |
+
"grad_norm": 0.8063498735427856,
|
| 932 |
+
"learning_rate": 5e-05,
|
| 933 |
+
"loss": 1.293,
|
| 934 |
+
"step": 131
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.04608134054808867,
|
| 938 |
+
"grad_norm": 1.117038369178772,
|
| 939 |
+
"learning_rate": 5e-05,
|
| 940 |
+
"loss": 1.3635,
|
| 941 |
+
"step": 132
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.04643044161284692,
|
| 945 |
+
"grad_norm": 1.4540647268295288,
|
| 946 |
+
"learning_rate": 5e-05,
|
| 947 |
+
"loss": 1.3346,
|
| 948 |
+
"step": 133
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.04677954267760517,
|
| 952 |
+
"grad_norm": 0.6695774793624878,
|
| 953 |
+
"learning_rate": 5e-05,
|
| 954 |
+
"loss": 1.4109,
|
| 955 |
+
"step": 134
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.04712864374236341,
|
| 959 |
+
"grad_norm": 0.8146533370018005,
|
| 960 |
+
"learning_rate": 5e-05,
|
| 961 |
+
"loss": 1.3515,
|
| 962 |
+
"step": 135
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.04747774480712166,
|
| 966 |
+
"grad_norm": 0.6705998778343201,
|
| 967 |
+
"learning_rate": 5e-05,
|
| 968 |
+
"loss": 1.2752,
|
| 969 |
+
"step": 136
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.04782684587187991,
|
| 973 |
+
"grad_norm": 0.7589219808578491,
|
| 974 |
+
"learning_rate": 5e-05,
|
| 975 |
+
"loss": 1.4393,
|
| 976 |
+
"step": 137
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.04817594693663816,
|
| 980 |
+
"grad_norm": 0.9603825807571411,
|
| 981 |
+
"learning_rate": 5e-05,
|
| 982 |
+
"loss": 1.4609,
|
| 983 |
+
"step": 138
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.048525048001396405,
|
| 987 |
+
"grad_norm": 0.6351510286331177,
|
| 988 |
+
"learning_rate": 5e-05,
|
| 989 |
+
"loss": 1.371,
|
| 990 |
+
"step": 139
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.04887414906615465,
|
| 994 |
+
"grad_norm": 0.5652881860733032,
|
| 995 |
+
"learning_rate": 5e-05,
|
| 996 |
+
"loss": 1.2845,
|
| 997 |
+
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.0492232501309129,
|
| 1001 |
+
"grad_norm": 0.7579118609428406,
|
| 1002 |
+
"learning_rate": 5e-05,
|
| 1003 |
+
"loss": 1.2526,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.04957235119567115,
|
| 1008 |
+
"grad_norm": 0.7851598262786865,
|
| 1009 |
+
"learning_rate": 5e-05,
|
| 1010 |
+
"loss": 1.3379,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.04992145226042939,
|
| 1015 |
+
"grad_norm": 0.5865357518196106,
|
| 1016 |
+
"learning_rate": 5e-05,
|
| 1017 |
+
"loss": 1.4802,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.05027055332518764,
|
| 1022 |
+
"grad_norm": 1.3862611055374146,
|
| 1023 |
+
"learning_rate": 5e-05,
|
| 1024 |
+
"loss": 1.357,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.05061965438994589,
|
| 1029 |
+
"grad_norm": 0.6249399185180664,
|
| 1030 |
+
"learning_rate": 5e-05,
|
| 1031 |
+
"loss": 1.2587,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.05096875545470414,
|
| 1036 |
+
"grad_norm": 0.5966644883155823,
|
| 1037 |
+
"learning_rate": 5e-05,
|
| 1038 |
+
"loss": 1.3534,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.051317856519462385,
|
| 1043 |
+
"grad_norm": 0.6312971711158752,
|
| 1044 |
+
"learning_rate": 5e-05,
|
| 1045 |
+
"loss": 1.1815,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.05166695758422063,
|
| 1050 |
+
"grad_norm": 0.6539703011512756,
|
| 1051 |
+
"learning_rate": 5e-05,
|
| 1052 |
+
"loss": 1.3946,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.05201605864897888,
|
| 1057 |
+
"grad_norm": 0.8756076097488403,
|
| 1058 |
+
"learning_rate": 5e-05,
|
| 1059 |
+
"loss": 1.2384,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.052365159713737126,
|
| 1064 |
+
"grad_norm": 0.7149311304092407,
|
| 1065 |
+
"learning_rate": 5e-05,
|
| 1066 |
+
"loss": 1.2998,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.05271426077849537,
|
| 1071 |
+
"grad_norm": 0.79525226354599,
|
| 1072 |
+
"learning_rate": 5e-05,
|
| 1073 |
+
"loss": 1.3376,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.05306336184325362,
|
| 1078 |
+
"grad_norm": 0.6921191811561584,
|
| 1079 |
+
"learning_rate": 5e-05,
|
| 1080 |
+
"loss": 1.3461,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.05341246290801187,
|
| 1085 |
+
"grad_norm": 0.7444896697998047,
|
| 1086 |
+
"learning_rate": 5e-05,
|
| 1087 |
+
"loss": 1.4089,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.05376156397277012,
|
| 1092 |
+
"grad_norm": 0.6216670274734497,
|
| 1093 |
+
"learning_rate": 5e-05,
|
| 1094 |
+
"loss": 1.3402,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.054110665037528365,
|
| 1099 |
+
"grad_norm": 0.5917710661888123,
|
| 1100 |
+
"learning_rate": 5e-05,
|
| 1101 |
+
"loss": 1.3253,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.05445976610228661,
|
| 1106 |
+
"grad_norm": 0.8648408055305481,
|
| 1107 |
+
"learning_rate": 5e-05,
|
| 1108 |
+
"loss": 1.4447,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.05480886716704486,
|
| 1113 |
+
"grad_norm": 0.6752570271492004,
|
| 1114 |
+
"learning_rate": 5e-05,
|
| 1115 |
+
"loss": 1.3097,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.055157968231803106,
|
| 1120 |
+
"grad_norm": 0.5603750944137573,
|
| 1121 |
+
"learning_rate": 5e-05,
|
| 1122 |
+
"loss": 1.4177,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.05550706929656136,
|
| 1127 |
+
"grad_norm": 0.6317929029464722,
|
| 1128 |
+
"learning_rate": 5e-05,
|
| 1129 |
+
"loss": 1.3509,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.0558561703613196,
|
| 1134 |
+
"grad_norm": 0.6017687320709229,
|
| 1135 |
+
"learning_rate": 5e-05,
|
| 1136 |
+
"loss": 1.3471,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.05620527142607785,
|
| 1141 |
+
"grad_norm": 0.6761009693145752,
|
| 1142 |
+
"learning_rate": 5e-05,
|
| 1143 |
+
"loss": 1.4473,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.0565543724908361,
|
| 1148 |
+
"grad_norm": 0.7266319990158081,
|
| 1149 |
+
"learning_rate": 5e-05,
|
| 1150 |
+
"loss": 1.2896,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.056903473555594344,
|
| 1155 |
+
"grad_norm": 0.6436321139335632,
|
| 1156 |
+
"learning_rate": 5e-05,
|
| 1157 |
+
"loss": 1.2812,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.05725257462035259,
|
| 1162 |
+
"grad_norm": 0.9664864540100098,
|
| 1163 |
+
"learning_rate": 5e-05,
|
| 1164 |
+
"loss": 1.294,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.05760167568511084,
|
| 1169 |
+
"grad_norm": 0.6690096855163574,
|
| 1170 |
+
"learning_rate": 5e-05,
|
| 1171 |
+
"loss": 1.2801,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.057950776749869086,
|
| 1176 |
+
"grad_norm": 0.6227753162384033,
|
| 1177 |
+
"learning_rate": 5e-05,
|
| 1178 |
+
"loss": 1.3384,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.05829987781462734,
|
| 1183 |
+
"grad_norm": 0.7900117039680481,
|
| 1184 |
+
"learning_rate": 5e-05,
|
| 1185 |
+
"loss": 1.3424,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.05864897887938558,
|
| 1190 |
+
"grad_norm": 0.6928064823150635,
|
| 1191 |
+
"learning_rate": 5e-05,
|
| 1192 |
+
"loss": 1.296,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.05899807994414383,
|
| 1197 |
+
"grad_norm": 0.8754634261131287,
|
| 1198 |
+
"learning_rate": 5e-05,
|
| 1199 |
+
"loss": 1.4471,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.05934718100890208,
|
| 1204 |
+
"grad_norm": 0.5537067651748657,
|
| 1205 |
+
"learning_rate": 5e-05,
|
| 1206 |
+
"loss": 1.2825,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.059696282073660324,
|
| 1211 |
+
"grad_norm": 0.6705783009529114,
|
| 1212 |
+
"learning_rate": 5e-05,
|
| 1213 |
+
"loss": 1.3768,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.06004538313841857,
|
| 1218 |
+
"grad_norm": 0.5732744932174683,
|
| 1219 |
+
"learning_rate": 5e-05,
|
| 1220 |
+
"loss": 1.3309,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.06039448420317682,
|
| 1225 |
+
"grad_norm": 1.120721459388733,
|
| 1226 |
+
"learning_rate": 5e-05,
|
| 1227 |
+
"loss": 1.3702,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.060743585267935066,
|
| 1232 |
+
"grad_norm": 0.7755718231201172,
|
| 1233 |
+
"learning_rate": 5e-05,
|
| 1234 |
+
"loss": 1.3425,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.06109268633269332,
|
| 1239 |
+
"grad_norm": 0.5984740257263184,
|
| 1240 |
+
"learning_rate": 5e-05,
|
| 1241 |
+
"loss": 1.4886,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.06144178739745156,
|
| 1246 |
+
"grad_norm": 0.7374542951583862,
|
| 1247 |
+
"learning_rate": 5e-05,
|
| 1248 |
+
"loss": 1.3667,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.06179088846220981,
|
| 1253 |
+
"grad_norm": 0.5558515787124634,
|
| 1254 |
+
"learning_rate": 5e-05,
|
| 1255 |
+
"loss": 1.3737,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.06213998952696806,
|
| 1260 |
+
"grad_norm": 0.700268566608429,
|
| 1261 |
+
"learning_rate": 5e-05,
|
| 1262 |
+
"loss": 1.364,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.062489090591726304,
|
| 1267 |
+
"grad_norm": 0.5781232118606567,
|
| 1268 |
+
"learning_rate": 5e-05,
|
| 1269 |
+
"loss": 1.3443,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.06283819165648455,
|
| 1274 |
+
"grad_norm": 0.7157448530197144,
|
| 1275 |
+
"learning_rate": 5e-05,
|
| 1276 |
+
"loss": 1.3702,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.0631872927212428,
|
| 1281 |
+
"grad_norm": 0.5329631567001343,
|
| 1282 |
+
"learning_rate": 5e-05,
|
| 1283 |
+
"loss": 1.1786,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.06353639378600105,
|
| 1288 |
+
"grad_norm": 0.5949011445045471,
|
| 1289 |
+
"learning_rate": 5e-05,
|
| 1290 |
+
"loss": 1.3809,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.0638854948507593,
|
| 1295 |
+
"grad_norm": 0.6756107807159424,
|
| 1296 |
+
"learning_rate": 5e-05,
|
| 1297 |
+
"loss": 1.2792,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.06423459591551754,
|
| 1302 |
+
"grad_norm": 0.7747790813446045,
|
| 1303 |
+
"learning_rate": 5e-05,
|
| 1304 |
+
"loss": 1.3714,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.06458369698027579,
|
| 1309 |
+
"grad_norm": 1.1907461881637573,
|
| 1310 |
+
"learning_rate": 5e-05,
|
| 1311 |
+
"loss": 1.3055,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.06493279804503403,
|
| 1316 |
+
"grad_norm": 0.5747818946838379,
|
| 1317 |
+
"learning_rate": 5e-05,
|
| 1318 |
+
"loss": 1.2003,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.06528189910979229,
|
| 1323 |
+
"grad_norm": 0.614464521408081,
|
| 1324 |
+
"learning_rate": 5e-05,
|
| 1325 |
+
"loss": 1.3108,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.06563100017455054,
|
| 1330 |
+
"grad_norm": 0.6040724515914917,
|
| 1331 |
+
"learning_rate": 5e-05,
|
| 1332 |
+
"loss": 1.2371,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.06598010123930878,
|
| 1337 |
+
"grad_norm": 0.6369174122810364,
|
| 1338 |
+
"learning_rate": 5e-05,
|
| 1339 |
+
"loss": 1.1662,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.06632920230406703,
|
| 1344 |
+
"grad_norm": 0.6132228374481201,
|
| 1345 |
+
"learning_rate": 5e-05,
|
| 1346 |
+
"loss": 1.3257,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.06667830336882527,
|
| 1351 |
+
"grad_norm": 0.6686124801635742,
|
| 1352 |
+
"learning_rate": 5e-05,
|
| 1353 |
+
"loss": 1.3757,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.06702740443358353,
|
| 1358 |
+
"grad_norm": 0.6709855794906616,
|
| 1359 |
+
"learning_rate": 5e-05,
|
| 1360 |
+
"loss": 1.3341,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.06737650549834177,
|
| 1365 |
+
"grad_norm": 0.5295905470848083,
|
| 1366 |
+
"learning_rate": 5e-05,
|
| 1367 |
+
"loss": 1.2587,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.06772560656310002,
|
| 1372 |
+
"grad_norm": 0.6111523509025574,
|
| 1373 |
+
"learning_rate": 5e-05,
|
| 1374 |
+
"loss": 1.3365,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.06807470762785826,
|
| 1379 |
+
"grad_norm": 0.5655878782272339,
|
| 1380 |
+
"learning_rate": 5e-05,
|
| 1381 |
+
"loss": 1.3265,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.06842380869261651,
|
| 1386 |
+
"grad_norm": 0.6125257015228271,
|
| 1387 |
+
"learning_rate": 5e-05,
|
| 1388 |
+
"loss": 1.3475,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.06877290975737475,
|
| 1393 |
+
"grad_norm": 0.6268573999404907,
|
| 1394 |
+
"learning_rate": 5e-05,
|
| 1395 |
+
"loss": 1.3002,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.06912201082213301,
|
| 1400 |
+
"grad_norm": 0.7267619967460632,
|
| 1401 |
+
"learning_rate": 5e-05,
|
| 1402 |
+
"loss": 1.4104,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.06947111188689126,
|
| 1407 |
+
"grad_norm": 0.5741710066795349,
|
| 1408 |
+
"learning_rate": 5e-05,
|
| 1409 |
+
"loss": 1.318,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.0698202129516495,
|
| 1414 |
+
"grad_norm": 0.6447280049324036,
|
| 1415 |
+
"learning_rate": 5e-05,
|
| 1416 |
+
"loss": 1.3477,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.0698202129516495,
|
| 1421 |
+
"eval_loss": 1.3300124406814575,
|
| 1422 |
+
"eval_runtime": 3301.7334,
|
| 1423 |
+
"eval_samples_per_second": 6.941,
|
| 1424 |
+
"eval_steps_per_second": 0.868,
|
| 1425 |
+
"step": 200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.07016931401640775,
|
| 1429 |
+
"grad_norm": 1.4164685010910034,
|
| 1430 |
+
"learning_rate": 5e-05,
|
| 1431 |
+
"loss": 1.4048,
|
| 1432 |
+
"step": 201
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 0.07051841508116599,
|
| 1436 |
+
"grad_norm": 0.5867809057235718,
|
| 1437 |
+
"learning_rate": 5e-05,
|
| 1438 |
+
"loss": 1.4018,
|
| 1439 |
+
"step": 202
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.07086751614592425,
|
| 1443 |
+
"grad_norm": 0.6882596611976624,
|
| 1444 |
+
"learning_rate": 5e-05,
|
| 1445 |
+
"loss": 1.2737,
|
| 1446 |
+
"step": 203
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.0712166172106825,
|
| 1450 |
+
"grad_norm": 0.6038634181022644,
|
| 1451 |
+
"learning_rate": 5e-05,
|
| 1452 |
+
"loss": 1.2399,
|
| 1453 |
+
"step": 204
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.07156571827544074,
|
| 1457 |
+
"grad_norm": 0.6428863406181335,
|
| 1458 |
+
"learning_rate": 5e-05,
|
| 1459 |
+
"loss": 1.3729,
|
| 1460 |
+
"step": 205
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.07191481934019898,
|
| 1464 |
+
"grad_norm": 0.7008076906204224,
|
| 1465 |
+
"learning_rate": 5e-05,
|
| 1466 |
+
"loss": 1.3353,
|
| 1467 |
+
"step": 206
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.07226392040495723,
|
| 1471 |
+
"grad_norm": 0.6662419438362122,
|
| 1472 |
+
"learning_rate": 5e-05,
|
| 1473 |
+
"loss": 1.3442,
|
| 1474 |
+
"step": 207
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.07261302146971549,
|
| 1478 |
+
"grad_norm": 0.7249788045883179,
|
| 1479 |
+
"learning_rate": 5e-05,
|
| 1480 |
+
"loss": 1.2526,
|
| 1481 |
+
"step": 208
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.07296212253447373,
|
| 1485 |
+
"grad_norm": 0.6323925852775574,
|
| 1486 |
+
"learning_rate": 5e-05,
|
| 1487 |
+
"loss": 1.2929,
|
| 1488 |
+
"step": 209
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.07331122359923198,
|
| 1492 |
+
"grad_norm": 0.8273724317550659,
|
| 1493 |
+
"learning_rate": 5e-05,
|
| 1494 |
+
"loss": 1.5291,
|
| 1495 |
+
"step": 210
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.07366032466399022,
|
| 1499 |
+
"grad_norm": 0.8445104956626892,
|
| 1500 |
+
"learning_rate": 5e-05,
|
| 1501 |
+
"loss": 1.2417,
|
| 1502 |
+
"step": 211
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.07400942572874847,
|
| 1506 |
+
"grad_norm": 0.6157236695289612,
|
| 1507 |
+
"learning_rate": 5e-05,
|
| 1508 |
+
"loss": 1.3739,
|
| 1509 |
+
"step": 212
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.07435852679350673,
|
| 1513 |
+
"grad_norm": 0.6917769312858582,
|
| 1514 |
+
"learning_rate": 5e-05,
|
| 1515 |
+
"loss": 1.3078,
|
| 1516 |
+
"step": 213
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.07470762785826497,
|
| 1520 |
+
"grad_norm": 0.7838917970657349,
|
| 1521 |
+
"learning_rate": 5e-05,
|
| 1522 |
+
"loss": 1.3086,
|
| 1523 |
+
"step": 214
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 0.07505672892302322,
|
| 1527 |
+
"grad_norm": 0.6962039470672607,
|
| 1528 |
+
"learning_rate": 5e-05,
|
| 1529 |
+
"loss": 1.3907,
|
| 1530 |
+
"step": 215
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.07540582998778146,
|
| 1534 |
+
"grad_norm": 0.6962039470672607,
|
| 1535 |
+
"learning_rate": 5e-05,
|
| 1536 |
+
"loss": 1.3615,
|
| 1537 |
+
"step": 216
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.0757549310525397,
|
| 1541 |
+
"grad_norm": 0.6687365770339966,
|
| 1542 |
+
"learning_rate": 5e-05,
|
| 1543 |
+
"loss": 1.3408,
|
| 1544 |
+
"step": 217
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.07610403211729795,
|
| 1548 |
+
"grad_norm": 0.5566404461860657,
|
| 1549 |
+
"learning_rate": 5e-05,
|
| 1550 |
+
"loss": 1.2872,
|
| 1551 |
+
"step": 218
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 0.07645313318205621,
|
| 1555 |
+
"grad_norm": 0.6419705748558044,
|
| 1556 |
+
"learning_rate": 5e-05,
|
| 1557 |
+
"loss": 1.2883,
|
| 1558 |
+
"step": 219
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 0.07680223424681445,
|
| 1562 |
+
"grad_norm": 0.7758398652076721,
|
| 1563 |
+
"learning_rate": 5e-05,
|
| 1564 |
+
"loss": 1.3832,
|
| 1565 |
+
"step": 220
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 0.0771513353115727,
|
| 1569 |
+
"grad_norm": 0.9763804078102112,
|
| 1570 |
+
"learning_rate": 5e-05,
|
| 1571 |
+
"loss": 1.3414,
|
| 1572 |
+
"step": 221
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.07750043637633094,
|
| 1576 |
+
"grad_norm": 0.8815904259681702,
|
| 1577 |
+
"learning_rate": 5e-05,
|
| 1578 |
+
"loss": 1.3297,
|
| 1579 |
+
"step": 222
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 0.07784953744108919,
|
| 1583 |
+
"grad_norm": 0.590263307094574,
|
| 1584 |
+
"learning_rate": 5e-05,
|
| 1585 |
+
"loss": 1.3401,
|
| 1586 |
+
"step": 223
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.07819863850584745,
|
| 1590 |
+
"grad_norm": 0.677057147026062,
|
| 1591 |
+
"learning_rate": 5e-05,
|
| 1592 |
+
"loss": 1.2449,
|
| 1593 |
+
"step": 224
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.07854773957060569,
|
| 1597 |
+
"grad_norm": 1.5185271501541138,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 1.3127,
|
| 1600 |
+
"step": 225
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.07889684063536394,
|
| 1604 |
+
"grad_norm": 0.5751495957374573,
|
| 1605 |
+
"learning_rate": 5e-05,
|
| 1606 |
+
"loss": 1.1587,
|
| 1607 |
+
"step": 226
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.07924594170012218,
|
| 1611 |
+
"grad_norm": 0.8122138977050781,
|
| 1612 |
+
"learning_rate": 5e-05,
|
| 1613 |
+
"loss": 1.2316,
|
| 1614 |
+
"step": 227
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.07959504276488043,
|
| 1618 |
+
"grad_norm": 0.6675130724906921,
|
| 1619 |
+
"learning_rate": 5e-05,
|
| 1620 |
+
"loss": 1.3539,
|
| 1621 |
+
"step": 228
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.07994414382963869,
|
| 1625 |
+
"grad_norm": 0.8163532614707947,
|
| 1626 |
+
"learning_rate": 5e-05,
|
| 1627 |
+
"loss": 1.328,
|
| 1628 |
+
"step": 229
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.08029324489439693,
|
| 1632 |
+
"grad_norm": 0.8377723693847656,
|
| 1633 |
+
"learning_rate": 5e-05,
|
| 1634 |
+
"loss": 1.353,
|
| 1635 |
+
"step": 230
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.08064234595915518,
|
| 1639 |
+
"grad_norm": 0.7325611710548401,
|
| 1640 |
+
"learning_rate": 5e-05,
|
| 1641 |
+
"loss": 1.3396,
|
| 1642 |
+
"step": 231
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 0.08099144702391342,
|
| 1646 |
+
"grad_norm": 0.8941824436187744,
|
| 1647 |
+
"learning_rate": 5e-05,
|
| 1648 |
+
"loss": 1.2906,
|
| 1649 |
+
"step": 232
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 0.08134054808867167,
|
| 1653 |
+
"grad_norm": 0.6284440159797668,
|
| 1654 |
+
"learning_rate": 5e-05,
|
| 1655 |
+
"loss": 1.4264,
|
| 1656 |
+
"step": 233
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 0.08168964915342992,
|
| 1660 |
+
"grad_norm": 0.689984917640686,
|
| 1661 |
+
"learning_rate": 5e-05,
|
| 1662 |
+
"loss": 1.3696,
|
| 1663 |
+
"step": 234
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 0.08203875021818817,
|
| 1667 |
+
"grad_norm": 0.5813177227973938,
|
| 1668 |
+
"learning_rate": 5e-05,
|
| 1669 |
+
"loss": 1.2931,
|
| 1670 |
+
"step": 235
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.08238785128294641,
|
| 1674 |
+
"grad_norm": 0.5287997126579285,
|
| 1675 |
+
"learning_rate": 5e-05,
|
| 1676 |
+
"loss": 1.3264,
|
| 1677 |
+
"step": 236
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 0.08273695234770466,
|
| 1681 |
+
"grad_norm": 0.7944268584251404,
|
| 1682 |
+
"learning_rate": 5e-05,
|
| 1683 |
+
"loss": 1.2708,
|
| 1684 |
+
"step": 237
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 0.0830860534124629,
|
| 1688 |
+
"grad_norm": 0.534864068031311,
|
| 1689 |
+
"learning_rate": 5e-05,
|
| 1690 |
+
"loss": 1.2535,
|
| 1691 |
+
"step": 238
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 0.08343515447722115,
|
| 1695 |
+
"grad_norm": 0.6260988712310791,
|
| 1696 |
+
"learning_rate": 5e-05,
|
| 1697 |
+
"loss": 1.2757,
|
| 1698 |
+
"step": 239
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 0.08378425554197941,
|
| 1702 |
+
"grad_norm": 0.579078197479248,
|
| 1703 |
+
"learning_rate": 5e-05,
|
| 1704 |
+
"loss": 1.2906,
|
| 1705 |
+
"step": 240
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.08413335660673765,
|
| 1709 |
+
"grad_norm": 0.5578561425209045,
|
| 1710 |
+
"learning_rate": 5e-05,
|
| 1711 |
+
"loss": 1.289,
|
| 1712 |
+
"step": 241
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.0844824576714959,
|
| 1716 |
+
"grad_norm": 0.626961350440979,
|
| 1717 |
+
"learning_rate": 5e-05,
|
| 1718 |
+
"loss": 1.2807,
|
| 1719 |
+
"step": 242
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 0.08483155873625414,
|
| 1723 |
+
"grad_norm": 0.782669186592102,
|
| 1724 |
+
"learning_rate": 5e-05,
|
| 1725 |
+
"loss": 1.3933,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 0.08518065980101239,
|
| 1730 |
+
"grad_norm": 0.6670363545417786,
|
| 1731 |
+
"learning_rate": 5e-05,
|
| 1732 |
+
"loss": 1.2732,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 0.08552976086577065,
|
| 1737 |
+
"grad_norm": 0.7201350331306458,
|
| 1738 |
+
"learning_rate": 5e-05,
|
| 1739 |
+
"loss": 1.2962,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.08587886193052889,
|
| 1744 |
+
"grad_norm": 0.6021212339401245,
|
| 1745 |
+
"learning_rate": 5e-05,
|
| 1746 |
+
"loss": 1.35,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 0.08622796299528714,
|
| 1751 |
+
"grad_norm": 0.8081540465354919,
|
| 1752 |
+
"learning_rate": 5e-05,
|
| 1753 |
+
"loss": 1.3568,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.08657706406004538,
|
| 1758 |
+
"grad_norm": 0.5358250737190247,
|
| 1759 |
+
"learning_rate": 5e-05,
|
| 1760 |
+
"loss": 1.4603,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 0.08692616512480363,
|
| 1765 |
+
"grad_norm": 0.6927733421325684,
|
| 1766 |
+
"learning_rate": 5e-05,
|
| 1767 |
+
"loss": 1.2506,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 0.08727526618956188,
|
| 1772 |
+
"grad_norm": 0.6187159419059753,
|
| 1773 |
+
"learning_rate": 5e-05,
|
| 1774 |
+
"loss": 1.3497,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 0.08762436725432013,
|
| 1779 |
+
"grad_norm": 0.6304159760475159,
|
| 1780 |
+
"learning_rate": 5e-05,
|
| 1781 |
+
"loss": 1.3087,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 0.08797346831907837,
|
| 1786 |
+
"grad_norm": 0.6446660161018372,
|
| 1787 |
+
"learning_rate": 5e-05,
|
| 1788 |
+
"loss": 1.3424,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 0.08832256938383662,
|
| 1793 |
+
"grad_norm": 0.6535473465919495,
|
| 1794 |
+
"learning_rate": 5e-05,
|
| 1795 |
+
"loss": 1.3471,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.08867167044859486,
|
| 1800 |
+
"grad_norm": 0.601290225982666,
|
| 1801 |
+
"learning_rate": 5e-05,
|
| 1802 |
+
"loss": 1.3557,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 0.08902077151335312,
|
| 1807 |
+
"grad_norm": 0.641854465007782,
|
| 1808 |
+
"learning_rate": 5e-05,
|
| 1809 |
+
"loss": 1.3138,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.08936987257811137,
|
| 1814 |
+
"grad_norm": 0.5452507138252258,
|
| 1815 |
+
"learning_rate": 5e-05,
|
| 1816 |
+
"loss": 1.2898,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.08971897364286961,
|
| 1821 |
+
"grad_norm": 0.5870373249053955,
|
| 1822 |
+
"learning_rate": 5e-05,
|
| 1823 |
+
"loss": 1.2953,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.09006807470762786,
|
| 1828 |
+
"grad_norm": 0.5798627734184265,
|
| 1829 |
+
"learning_rate": 5e-05,
|
| 1830 |
+
"loss": 1.2973,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.0904171757723861,
|
| 1835 |
+
"grad_norm": 0.5798627734184265,
|
| 1836 |
+
"learning_rate": 5e-05,
|
| 1837 |
+
"loss": 1.3628,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.09076627683714435,
|
| 1842 |
+
"grad_norm": 0.7382280230522156,
|
| 1843 |
+
"learning_rate": 5e-05,
|
| 1844 |
+
"loss": 1.3111,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.0911153779019026,
|
| 1849 |
+
"grad_norm": 0.6882988810539246,
|
| 1850 |
+
"learning_rate": 5e-05,
|
| 1851 |
+
"loss": 1.329,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 0.09146447896666085,
|
| 1856 |
+
"grad_norm": 0.6590788960456848,
|
| 1857 |
+
"learning_rate": 5e-05,
|
| 1858 |
+
"loss": 1.3089,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 0.0918135800314191,
|
| 1863 |
+
"grad_norm": 0.682006299495697,
|
| 1864 |
+
"learning_rate": 5e-05,
|
| 1865 |
+
"loss": 1.344,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.09216268109617734,
|
| 1870 |
+
"grad_norm": 0.6040222644805908,
|
| 1871 |
+
"learning_rate": 5e-05,
|
| 1872 |
+
"loss": 1.3919,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.09251178216093559,
|
| 1877 |
+
"grad_norm": 0.5964936017990112,
|
| 1878 |
+
"learning_rate": 5e-05,
|
| 1879 |
+
"loss": 1.3397,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.09286088322569384,
|
| 1884 |
+
"grad_norm": 0.5645217299461365,
|
| 1885 |
+
"learning_rate": 5e-05,
|
| 1886 |
+
"loss": 1.3488,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 0.09320998429045209,
|
| 1891 |
+
"grad_norm": 0.7771989703178406,
|
| 1892 |
+
"learning_rate": 5e-05,
|
| 1893 |
+
"loss": 1.3485,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 0.09355908535521033,
|
| 1898 |
+
"grad_norm": 0.6003885865211487,
|
| 1899 |
+
"learning_rate": 5e-05,
|
| 1900 |
+
"loss": 1.3109,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.09390818641996858,
|
| 1905 |
+
"grad_norm": 0.5627903938293457,
|
| 1906 |
+
"learning_rate": 5e-05,
|
| 1907 |
+
"loss": 1.2906,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 0.09425728748472682,
|
| 1912 |
+
"grad_norm": 0.6381875276565552,
|
| 1913 |
+
"learning_rate": 5e-05,
|
| 1914 |
+
"loss": 1.3063,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.09460638854948508,
|
| 1919 |
+
"grad_norm": 1.2558772563934326,
|
| 1920 |
+
"learning_rate": 5e-05,
|
| 1921 |
+
"loss": 1.2985,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.09495548961424333,
|
| 1926 |
+
"grad_norm": 0.6977007389068604,
|
| 1927 |
+
"learning_rate": 5e-05,
|
| 1928 |
+
"loss": 1.4955,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.09530459067900157,
|
| 1933 |
+
"grad_norm": 0.7846536040306091,
|
| 1934 |
+
"learning_rate": 5e-05,
|
| 1935 |
+
"loss": 1.4439,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.09565369174375982,
|
| 1940 |
+
"grad_norm": 0.7036994695663452,
|
| 1941 |
+
"learning_rate": 5e-05,
|
| 1942 |
+
"loss": 1.1942,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.09600279280851806,
|
| 1947 |
+
"grad_norm": 0.6119917631149292,
|
| 1948 |
+
"learning_rate": 5e-05,
|
| 1949 |
+
"loss": 1.3607,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.09635189387327632,
|
| 1954 |
+
"grad_norm": 0.6243535280227661,
|
| 1955 |
+
"learning_rate": 5e-05,
|
| 1956 |
+
"loss": 1.3029,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 0.09670099493803457,
|
| 1961 |
+
"grad_norm": 0.5424296855926514,
|
| 1962 |
+
"learning_rate": 5e-05,
|
| 1963 |
+
"loss": 1.2995,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.09705009600279281,
|
| 1968 |
+
"grad_norm": 0.7677564024925232,
|
| 1969 |
+
"learning_rate": 5e-05,
|
| 1970 |
+
"loss": 1.2686,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 0.09739919706755105,
|
| 1975 |
+
"grad_norm": 0.625275194644928,
|
| 1976 |
+
"learning_rate": 5e-05,
|
| 1977 |
+
"loss": 1.2897,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 0.0977482981323093,
|
| 1982 |
+
"grad_norm": 0.5734910368919373,
|
| 1983 |
+
"learning_rate": 5e-05,
|
| 1984 |
+
"loss": 1.3298,
|
| 1985 |
+
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 0.09809739919706754,
|
| 1989 |
+
"grad_norm": 0.660658061504364,
|
| 1990 |
+
"learning_rate": 5e-05,
|
| 1991 |
+
"loss": 1.2643,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.0984465002618258,
|
| 1996 |
+
"grad_norm": 0.679891049861908,
|
| 1997 |
+
"learning_rate": 5e-05,
|
| 1998 |
+
"loss": 1.3189,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.09879560132658405,
|
| 2003 |
+
"grad_norm": 0.6248694658279419,
|
| 2004 |
+
"learning_rate": 5e-05,
|
| 2005 |
+
"loss": 1.1688,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.0991447023913423,
|
| 2010 |
+
"grad_norm": 0.6428897380828857,
|
| 2011 |
+
"learning_rate": 5e-05,
|
| 2012 |
+
"loss": 1.3274,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 0.09949380345610054,
|
| 2017 |
+
"grad_norm": 0.586065411567688,
|
| 2018 |
+
"learning_rate": 5e-05,
|
| 2019 |
+
"loss": 1.3852,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.09984290452085878,
|
| 2024 |
+
"grad_norm": 0.5755594372749329,
|
| 2025 |
+
"learning_rate": 5e-05,
|
| 2026 |
+
"loss": 1.3665,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.10019200558561704,
|
| 2031 |
+
"grad_norm": 0.7748963236808777,
|
| 2032 |
+
"learning_rate": 5e-05,
|
| 2033 |
+
"loss": 1.4551,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.10054110665037529,
|
| 2038 |
+
"grad_norm": 0.6308531165122986,
|
| 2039 |
+
"learning_rate": 5e-05,
|
| 2040 |
+
"loss": 1.2793,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.10089020771513353,
|
| 2045 |
+
"grad_norm": 0.6195006966590881,
|
| 2046 |
+
"learning_rate": 5e-05,
|
| 2047 |
+
"loss": 1.3649,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.10123930877989178,
|
| 2052 |
+
"grad_norm": 0.6098636984825134,
|
| 2053 |
+
"learning_rate": 5e-05,
|
| 2054 |
+
"loss": 1.2956,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 0.10158840984465002,
|
| 2059 |
+
"grad_norm": 0.8072320818901062,
|
| 2060 |
+
"learning_rate": 5e-05,
|
| 2061 |
+
"loss": 1.3469,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.10193751090940828,
|
| 2066 |
+
"grad_norm": 0.6090126633644104,
|
| 2067 |
+
"learning_rate": 5e-05,
|
| 2068 |
+
"loss": 1.2958,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 0.10228661197416652,
|
| 2073 |
+
"grad_norm": 0.5718780159950256,
|
| 2074 |
+
"learning_rate": 5e-05,
|
| 2075 |
+
"loss": 1.363,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 0.10263571303892477,
|
| 2080 |
+
"grad_norm": 0.7197532653808594,
|
| 2081 |
+
"learning_rate": 5e-05,
|
| 2082 |
+
"loss": 1.3868,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 0.10298481410368301,
|
| 2087 |
+
"grad_norm": 0.5578592419624329,
|
| 2088 |
+
"learning_rate": 5e-05,
|
| 2089 |
+
"loss": 1.2627,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.10333391516844126,
|
| 2094 |
+
"grad_norm": 0.730226457118988,
|
| 2095 |
+
"learning_rate": 5e-05,
|
| 2096 |
+
"loss": 1.3182,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 0.10368301623319952,
|
| 2101 |
+
"grad_norm": 0.6234796047210693,
|
| 2102 |
+
"learning_rate": 5e-05,
|
| 2103 |
+
"loss": 1.1777,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 0.10403211729795776,
|
| 2108 |
+
"grad_norm": 0.5563578009605408,
|
| 2109 |
+
"learning_rate": 5e-05,
|
| 2110 |
+
"loss": 1.3275,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.10438121836271601,
|
| 2115 |
+
"grad_norm": 0.6864249110221863,
|
| 2116 |
+
"learning_rate": 5e-05,
|
| 2117 |
+
"loss": 1.2813,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 0.10473031942747425,
|
| 2122 |
+
"grad_norm": 0.8850319385528564,
|
| 2123 |
+
"learning_rate": 5e-05,
|
| 2124 |
+
"loss": 1.3057,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 0.10473031942747425,
|
| 2129 |
+
"eval_loss": 1.3255380392074585,
|
| 2130 |
+
"eval_runtime": 3311.4237,
|
| 2131 |
+
"eval_samples_per_second": 6.92,
|
| 2132 |
+
"eval_steps_per_second": 0.865,
|
| 2133 |
+
"step": 300
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 0.1050794204922325,
|
| 2137 |
+
"grad_norm": 0.9439303278923035,
|
| 2138 |
+
"learning_rate": 5e-05,
|
| 2139 |
+
"loss": 1.281,
|
| 2140 |
+
"step": 301
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 0.10542852155699074,
|
| 2144 |
+
"grad_norm": 0.6651242971420288,
|
| 2145 |
+
"learning_rate": 5e-05,
|
| 2146 |
+
"loss": 1.3492,
|
| 2147 |
+
"step": 302
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 0.105777622621749,
|
| 2151 |
+
"grad_norm": 0.9047183394432068,
|
| 2152 |
+
"learning_rate": 5e-05,
|
| 2153 |
+
"loss": 1.4246,
|
| 2154 |
+
"step": 303
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 0.10612672368650725,
|
| 2158 |
+
"grad_norm": 0.6983138918876648,
|
| 2159 |
+
"learning_rate": 5e-05,
|
| 2160 |
+
"loss": 1.324,
|
| 2161 |
+
"step": 304
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 0.10647582475126549,
|
| 2165 |
+
"grad_norm": 0.6347063779830933,
|
| 2166 |
+
"learning_rate": 5e-05,
|
| 2167 |
+
"loss": 1.3389,
|
| 2168 |
+
"step": 305
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 0.10682492581602374,
|
| 2172 |
+
"grad_norm": 0.6051842570304871,
|
| 2173 |
+
"learning_rate": 5e-05,
|
| 2174 |
+
"loss": 1.3278,
|
| 2175 |
+
"step": 306
|
| 2176 |
+
},
|
| 2177 |
+
{
|
| 2178 |
+
"epoch": 0.10717402688078198,
|
| 2179 |
+
"grad_norm": 0.9355935454368591,
|
| 2180 |
+
"learning_rate": 5e-05,
|
| 2181 |
+
"loss": 1.2663,
|
| 2182 |
+
"step": 307
|
| 2183 |
+
},
|
| 2184 |
+
{
|
| 2185 |
+
"epoch": 0.10752312794554024,
|
| 2186 |
+
"grad_norm": 1.0706268548965454,
|
| 2187 |
+
"learning_rate": 5e-05,
|
| 2188 |
+
"loss": 1.3142,
|
| 2189 |
+
"step": 308
|
| 2190 |
+
},
|
| 2191 |
+
{
|
| 2192 |
+
"epoch": 0.10787222901029848,
|
| 2193 |
+
"grad_norm": 0.8131638765335083,
|
| 2194 |
+
"learning_rate": 5e-05,
|
| 2195 |
+
"loss": 1.3445,
|
| 2196 |
+
"step": 309
|
| 2197 |
+
},
|
| 2198 |
+
{
|
| 2199 |
+
"epoch": 0.10822133007505673,
|
| 2200 |
+
"grad_norm": 0.5791985392570496,
|
| 2201 |
+
"learning_rate": 5e-05,
|
| 2202 |
+
"loss": 1.2746,
|
| 2203 |
+
"step": 310
|
| 2204 |
+
},
|
| 2205 |
+
{
|
| 2206 |
+
"epoch": 0.10857043113981497,
|
| 2207 |
+
"grad_norm": 0.5536484718322754,
|
| 2208 |
+
"learning_rate": 5e-05,
|
| 2209 |
+
"loss": 1.2613,
|
| 2210 |
+
"step": 311
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 0.10891953220457322,
|
| 2214 |
+
"grad_norm": 0.7847089767456055,
|
| 2215 |
+
"learning_rate": 5e-05,
|
| 2216 |
+
"loss": 1.4607,
|
| 2217 |
+
"step": 312
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 0.10926863326933148,
|
| 2221 |
+
"grad_norm": 0.7828165888786316,
|
| 2222 |
+
"learning_rate": 5e-05,
|
| 2223 |
+
"loss": 1.4399,
|
| 2224 |
+
"step": 313
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 0.10961773433408972,
|
| 2228 |
+
"grad_norm": 0.5692522525787354,
|
| 2229 |
+
"learning_rate": 5e-05,
|
| 2230 |
+
"loss": 1.3044,
|
| 2231 |
+
"step": 314
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"epoch": 0.10996683539884797,
|
| 2235 |
+
"grad_norm": 0.5592648386955261,
|
| 2236 |
+
"learning_rate": 5e-05,
|
| 2237 |
+
"loss": 1.3211,
|
| 2238 |
+
"step": 315
|
| 2239 |
+
},
|
| 2240 |
+
{
|
| 2241 |
+
"epoch": 0.11031593646360621,
|
| 2242 |
+
"grad_norm": 0.7055444717407227,
|
| 2243 |
+
"learning_rate": 5e-05,
|
| 2244 |
+
"loss": 1.2944,
|
| 2245 |
+
"step": 316
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 0.11066503752836446,
|
| 2249 |
+
"grad_norm": 0.5370152592658997,
|
| 2250 |
+
"learning_rate": 5e-05,
|
| 2251 |
+
"loss": 1.2776,
|
| 2252 |
+
"step": 317
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 0.11101413859312272,
|
| 2256 |
+
"grad_norm": 0.6320214867591858,
|
| 2257 |
+
"learning_rate": 5e-05,
|
| 2258 |
+
"loss": 1.347,
|
| 2259 |
+
"step": 318
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 0.11136323965788096,
|
| 2263 |
+
"grad_norm": 0.6425771713256836,
|
| 2264 |
+
"learning_rate": 5e-05,
|
| 2265 |
+
"loss": 1.5038,
|
| 2266 |
+
"step": 319
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 0.1117123407226392,
|
| 2270 |
+
"grad_norm": 0.585542619228363,
|
| 2271 |
+
"learning_rate": 5e-05,
|
| 2272 |
+
"loss": 1.3573,
|
| 2273 |
+
"step": 320
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 0.11206144178739745,
|
| 2277 |
+
"grad_norm": 0.5627699494361877,
|
| 2278 |
+
"learning_rate": 5e-05,
|
| 2279 |
+
"loss": 1.2693,
|
| 2280 |
+
"step": 321
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 0.1124105428521557,
|
| 2284 |
+
"grad_norm": 0.6050506830215454,
|
| 2285 |
+
"learning_rate": 5e-05,
|
| 2286 |
+
"loss": 1.2787,
|
| 2287 |
+
"step": 322
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 0.11275964391691394,
|
| 2291 |
+
"grad_norm": 0.6247337460517883,
|
| 2292 |
+
"learning_rate": 5e-05,
|
| 2293 |
+
"loss": 1.4146,
|
| 2294 |
+
"step": 323
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 0.1131087449816722,
|
| 2298 |
+
"grad_norm": 0.7732966542243958,
|
| 2299 |
+
"learning_rate": 5e-05,
|
| 2300 |
+
"loss": 1.2626,
|
| 2301 |
+
"step": 324
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"epoch": 0.11345784604643044,
|
| 2305 |
+
"grad_norm": 0.5666255354881287,
|
| 2306 |
+
"learning_rate": 5e-05,
|
| 2307 |
+
"loss": 1.4219,
|
| 2308 |
+
"step": 325
|
| 2309 |
+
},
|
| 2310 |
+
{
|
| 2311 |
+
"epoch": 0.11380694711118869,
|
| 2312 |
+
"grad_norm": 0.5973132848739624,
|
| 2313 |
+
"learning_rate": 5e-05,
|
| 2314 |
+
"loss": 1.3522,
|
| 2315 |
+
"step": 326
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 0.11415604817594693,
|
| 2319 |
+
"grad_norm": 0.8540626764297485,
|
| 2320 |
+
"learning_rate": 5e-05,
|
| 2321 |
+
"loss": 1.304,
|
| 2322 |
+
"step": 327
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 0.11450514924070518,
|
| 2326 |
+
"grad_norm": 0.574573278427124,
|
| 2327 |
+
"learning_rate": 5e-05,
|
| 2328 |
+
"loss": 1.3487,
|
| 2329 |
+
"step": 328
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 0.11485425030546344,
|
| 2333 |
+
"grad_norm": 0.5949917435646057,
|
| 2334 |
+
"learning_rate": 5e-05,
|
| 2335 |
+
"loss": 1.254,
|
| 2336 |
+
"step": 329
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 0.11520335137022168,
|
| 2340 |
+
"grad_norm": 0.6005589365959167,
|
| 2341 |
+
"learning_rate": 5e-05,
|
| 2342 |
+
"loss": 1.3073,
|
| 2343 |
+
"step": 330
|
| 2344 |
+
},
|
| 2345 |
+
{
|
| 2346 |
+
"epoch": 0.11555245243497993,
|
| 2347 |
+
"grad_norm": 0.5026714205741882,
|
| 2348 |
+
"learning_rate": 5e-05,
|
| 2349 |
+
"loss": 1.2418,
|
| 2350 |
+
"step": 331
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 0.11590155349973817,
|
| 2354 |
+
"grad_norm": 0.7160278558731079,
|
| 2355 |
+
"learning_rate": 5e-05,
|
| 2356 |
+
"loss": 1.3437,
|
| 2357 |
+
"step": 332
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 0.11625065456449642,
|
| 2361 |
+
"grad_norm": 0.6049554347991943,
|
| 2362 |
+
"learning_rate": 5e-05,
|
| 2363 |
+
"loss": 1.4858,
|
| 2364 |
+
"step": 333
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 0.11659975562925468,
|
| 2368 |
+
"grad_norm": 0.7706385254859924,
|
| 2369 |
+
"learning_rate": 5e-05,
|
| 2370 |
+
"loss": 1.3971,
|
| 2371 |
+
"step": 334
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 0.11694885669401292,
|
| 2375 |
+
"grad_norm": 0.6254088282585144,
|
| 2376 |
+
"learning_rate": 5e-05,
|
| 2377 |
+
"loss": 1.3359,
|
| 2378 |
+
"step": 335
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 0.11729795775877117,
|
| 2382 |
+
"grad_norm": 0.5904930830001831,
|
| 2383 |
+
"learning_rate": 5e-05,
|
| 2384 |
+
"loss": 1.3262,
|
| 2385 |
+
"step": 336
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 0.11764705882352941,
|
| 2389 |
+
"grad_norm": 1.9982556104660034,
|
| 2390 |
+
"learning_rate": 5e-05,
|
| 2391 |
+
"loss": 1.3656,
|
| 2392 |
+
"step": 337
|
| 2393 |
+
},
|
| 2394 |
+
{
|
| 2395 |
+
"epoch": 0.11799615988828766,
|
| 2396 |
+
"grad_norm": 0.5776758790016174,
|
| 2397 |
+
"learning_rate": 5e-05,
|
| 2398 |
+
"loss": 1.2654,
|
| 2399 |
+
"step": 338
|
| 2400 |
+
},
|
| 2401 |
+
{
|
| 2402 |
+
"epoch": 0.1183452609530459,
|
| 2403 |
+
"grad_norm": 0.6094497442245483,
|
| 2404 |
+
"learning_rate": 5e-05,
|
| 2405 |
+
"loss": 1.3505,
|
| 2406 |
+
"step": 339
|
| 2407 |
+
},
|
| 2408 |
+
{
|
| 2409 |
+
"epoch": 0.11869436201780416,
|
| 2410 |
+
"grad_norm": 0.9940481185913086,
|
| 2411 |
+
"learning_rate": 5e-05,
|
| 2412 |
+
"loss": 1.2853,
|
| 2413 |
+
"step": 340
|
| 2414 |
+
},
|
| 2415 |
+
{
|
| 2416 |
+
"epoch": 0.1190434630825624,
|
| 2417 |
+
"grad_norm": 1.1043668985366821,
|
| 2418 |
+
"learning_rate": 5e-05,
|
| 2419 |
+
"loss": 1.2813,
|
| 2420 |
+
"step": 341
|
| 2421 |
+
},
|
| 2422 |
+
{
|
| 2423 |
+
"epoch": 0.11939256414732065,
|
| 2424 |
+
"grad_norm": 0.5494128465652466,
|
| 2425 |
+
"learning_rate": 5e-05,
|
| 2426 |
+
"loss": 1.202,
|
| 2427 |
+
"step": 342
|
| 2428 |
+
},
|
| 2429 |
+
{
|
| 2430 |
+
"epoch": 0.1197416652120789,
|
| 2431 |
+
"grad_norm": 0.6436132192611694,
|
| 2432 |
+
"learning_rate": 5e-05,
|
| 2433 |
+
"loss": 1.2898,
|
| 2434 |
+
"step": 343
|
| 2435 |
+
},
|
| 2436 |
+
{
|
| 2437 |
+
"epoch": 0.12009076627683714,
|
| 2438 |
+
"grad_norm": 0.6878450512886047,
|
| 2439 |
+
"learning_rate": 5e-05,
|
| 2440 |
+
"loss": 1.3392,
|
| 2441 |
+
"step": 344
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 0.1204398673415954,
|
| 2445 |
+
"grad_norm": 0.5806905627250671,
|
| 2446 |
+
"learning_rate": 5e-05,
|
| 2447 |
+
"loss": 1.2221,
|
| 2448 |
+
"step": 345
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 0.12078896840635364,
|
| 2452 |
+
"grad_norm": 0.5916112065315247,
|
| 2453 |
+
"learning_rate": 5e-05,
|
| 2454 |
+
"loss": 1.2761,
|
| 2455 |
+
"step": 346
|
| 2456 |
+
},
|
| 2457 |
+
{
|
| 2458 |
+
"epoch": 0.12113806947111189,
|
| 2459 |
+
"grad_norm": 0.5216647386550903,
|
| 2460 |
+
"learning_rate": 5e-05,
|
| 2461 |
+
"loss": 1.223,
|
| 2462 |
+
"step": 347
|
| 2463 |
+
},
|
| 2464 |
+
{
|
| 2465 |
+
"epoch": 0.12148717053587013,
|
| 2466 |
+
"grad_norm": 0.707747220993042,
|
| 2467 |
+
"learning_rate": 5e-05,
|
| 2468 |
+
"loss": 1.2933,
|
| 2469 |
+
"step": 348
|
| 2470 |
+
},
|
| 2471 |
+
{
|
| 2472 |
+
"epoch": 0.12183627160062838,
|
| 2473 |
+
"grad_norm": 0.6644443273544312,
|
| 2474 |
+
"learning_rate": 5e-05,
|
| 2475 |
+
"loss": 1.3367,
|
| 2476 |
+
"step": 349
|
| 2477 |
+
},
|
| 2478 |
+
{
|
| 2479 |
+
"epoch": 0.12218537266538664,
|
| 2480 |
+
"grad_norm": 0.7112720012664795,
|
| 2481 |
+
"learning_rate": 5e-05,
|
| 2482 |
+
"loss": 1.2368,
|
| 2483 |
+
"step": 350
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 0.12253447373014488,
|
| 2487 |
+
"grad_norm": 0.6551552414894104,
|
| 2488 |
+
"learning_rate": 5e-05,
|
| 2489 |
+
"loss": 1.3348,
|
| 2490 |
+
"step": 351
|
| 2491 |
+
},
|
| 2492 |
+
{
|
| 2493 |
+
"epoch": 0.12288357479490312,
|
| 2494 |
+
"grad_norm": 0.5377748012542725,
|
| 2495 |
+
"learning_rate": 5e-05,
|
| 2496 |
+
"loss": 1.2859,
|
| 2497 |
+
"step": 352
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"epoch": 0.12323267585966137,
|
| 2501 |
+
"grad_norm": 0.580769956111908,
|
| 2502 |
+
"learning_rate": 5e-05,
|
| 2503 |
+
"loss": 1.2442,
|
| 2504 |
+
"step": 353
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 0.12358177692441961,
|
| 2508 |
+
"grad_norm": 0.6772916316986084,
|
| 2509 |
+
"learning_rate": 5e-05,
|
| 2510 |
+
"loss": 1.2994,
|
| 2511 |
+
"step": 354
|
| 2512 |
+
},
|
| 2513 |
+
{
|
| 2514 |
+
"epoch": 0.12393087798917787,
|
| 2515 |
+
"grad_norm": 0.6245989799499512,
|
| 2516 |
+
"learning_rate": 5e-05,
|
| 2517 |
+
"loss": 1.2093,
|
| 2518 |
+
"step": 355
|
| 2519 |
+
},
|
| 2520 |
+
{
|
| 2521 |
+
"epoch": 0.12427997905393612,
|
| 2522 |
+
"grad_norm": 0.6136452555656433,
|
| 2523 |
+
"learning_rate": 5e-05,
|
| 2524 |
+
"loss": 1.2258,
|
| 2525 |
+
"step": 356
|
| 2526 |
+
},
|
| 2527 |
+
{
|
| 2528 |
+
"epoch": 0.12462908011869436,
|
| 2529 |
+
"grad_norm": 0.5786277055740356,
|
| 2530 |
+
"learning_rate": 5e-05,
|
| 2531 |
+
"loss": 1.2856,
|
| 2532 |
+
"step": 357
|
| 2533 |
+
},
|
| 2534 |
+
{
|
| 2535 |
+
"epoch": 0.12497818118345261,
|
| 2536 |
+
"grad_norm": 0.5986611247062683,
|
| 2537 |
+
"learning_rate": 5e-05,
|
| 2538 |
+
"loss": 1.4524,
|
| 2539 |
+
"step": 358
|
| 2540 |
+
},
|
| 2541 |
+
{
|
| 2542 |
+
"epoch": 0.12532728224821085,
|
| 2543 |
+
"grad_norm": 0.6240454316139221,
|
| 2544 |
+
"learning_rate": 5e-05,
|
| 2545 |
+
"loss": 1.3325,
|
| 2546 |
+
"step": 359
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 0.1256763833129691,
|
| 2550 |
+
"grad_norm": 0.6426084041595459,
|
| 2551 |
+
"learning_rate": 5e-05,
|
| 2552 |
+
"loss": 1.219,
|
| 2553 |
+
"step": 360
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 0.12602548437772734,
|
| 2557 |
+
"grad_norm": 0.6227401494979858,
|
| 2558 |
+
"learning_rate": 5e-05,
|
| 2559 |
+
"loss": 1.3342,
|
| 2560 |
+
"step": 361
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 0.1263745854424856,
|
| 2564 |
+
"grad_norm": 0.7462456226348877,
|
| 2565 |
+
"learning_rate": 5e-05,
|
| 2566 |
+
"loss": 1.3747,
|
| 2567 |
+
"step": 362
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 0.12672368650724386,
|
| 2571 |
+
"grad_norm": 0.7022641897201538,
|
| 2572 |
+
"learning_rate": 5e-05,
|
| 2573 |
+
"loss": 1.2957,
|
| 2574 |
+
"step": 363
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 0.1270727875720021,
|
| 2578 |
+
"grad_norm": 0.657645046710968,
|
| 2579 |
+
"learning_rate": 5e-05,
|
| 2580 |
+
"loss": 1.3125,
|
| 2581 |
+
"step": 364
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 0.12742188863676035,
|
| 2585 |
+
"grad_norm": 0.662497878074646,
|
| 2586 |
+
"learning_rate": 5e-05,
|
| 2587 |
+
"loss": 1.321,
|
| 2588 |
+
"step": 365
|
| 2589 |
+
},
|
| 2590 |
+
{
|
| 2591 |
+
"epoch": 0.1277709897015186,
|
| 2592 |
+
"grad_norm": 0.6295817494392395,
|
| 2593 |
+
"learning_rate": 5e-05,
|
| 2594 |
+
"loss": 1.3814,
|
| 2595 |
+
"step": 366
|
| 2596 |
+
},
|
| 2597 |
+
{
|
| 2598 |
+
"epoch": 0.12812009076627684,
|
| 2599 |
+
"grad_norm": 0.7357390522956848,
|
| 2600 |
+
"learning_rate": 5e-05,
|
| 2601 |
+
"loss": 1.374,
|
| 2602 |
+
"step": 367
|
| 2603 |
+
},
|
| 2604 |
+
{
|
| 2605 |
+
"epoch": 0.12846919183103508,
|
| 2606 |
+
"grad_norm": 0.6728739142417908,
|
| 2607 |
+
"learning_rate": 5e-05,
|
| 2608 |
+
"loss": 1.1957,
|
| 2609 |
+
"step": 368
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 0.12881829289579333,
|
| 2613 |
+
"grad_norm": 0.6290231943130493,
|
| 2614 |
+
"learning_rate": 5e-05,
|
| 2615 |
+
"loss": 1.2948,
|
| 2616 |
+
"step": 369
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 0.12916739396055157,
|
| 2620 |
+
"grad_norm": 1.0889554023742676,
|
| 2621 |
+
"learning_rate": 5e-05,
|
| 2622 |
+
"loss": 1.3465,
|
| 2623 |
+
"step": 370
|
| 2624 |
+
},
|
| 2625 |
+
{
|
| 2626 |
+
"epoch": 0.12951649502530982,
|
| 2627 |
+
"grad_norm": 0.6978388428688049,
|
| 2628 |
+
"learning_rate": 5e-05,
|
| 2629 |
+
"loss": 1.2898,
|
| 2630 |
+
"step": 371
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 0.12986559609006806,
|
| 2634 |
+
"grad_norm": 1.0806949138641357,
|
| 2635 |
+
"learning_rate": 5e-05,
|
| 2636 |
+
"loss": 1.2656,
|
| 2637 |
+
"step": 372
|
| 2638 |
+
},
|
| 2639 |
+
{
|
| 2640 |
+
"epoch": 0.1302146971548263,
|
| 2641 |
+
"grad_norm": 0.5989696979522705,
|
| 2642 |
+
"learning_rate": 5e-05,
|
| 2643 |
+
"loss": 1.354,
|
| 2644 |
+
"step": 373
|
| 2645 |
+
},
|
| 2646 |
+
{
|
| 2647 |
+
"epoch": 0.13056379821958458,
|
| 2648 |
+
"grad_norm": 0.5808868408203125,
|
| 2649 |
+
"learning_rate": 5e-05,
|
| 2650 |
+
"loss": 1.2911,
|
| 2651 |
+
"step": 374
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"epoch": 0.13091289928434283,
|
| 2655 |
+
"grad_norm": 0.6175510883331299,
|
| 2656 |
+
"learning_rate": 5e-05,
|
| 2657 |
+
"loss": 1.3392,
|
| 2658 |
+
"step": 375
|
| 2659 |
+
},
|
| 2660 |
+
{
|
| 2661 |
+
"epoch": 0.13126200034910107,
|
| 2662 |
+
"grad_norm": 0.7896063923835754,
|
| 2663 |
+
"learning_rate": 5e-05,
|
| 2664 |
+
"loss": 1.3598,
|
| 2665 |
+
"step": 376
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 0.13161110141385932,
|
| 2669 |
+
"grad_norm": 0.6890353560447693,
|
| 2670 |
+
"learning_rate": 5e-05,
|
| 2671 |
+
"loss": 1.2259,
|
| 2672 |
+
"step": 377
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 0.13196020247861756,
|
| 2676 |
+
"grad_norm": 0.7264868021011353,
|
| 2677 |
+
"learning_rate": 5e-05,
|
| 2678 |
+
"loss": 1.3747,
|
| 2679 |
+
"step": 378
|
| 2680 |
+
},
|
| 2681 |
+
{
|
| 2682 |
+
"epoch": 0.1323093035433758,
|
| 2683 |
+
"grad_norm": 0.5779114365577698,
|
| 2684 |
+
"learning_rate": 5e-05,
|
| 2685 |
+
"loss": 1.2566,
|
| 2686 |
+
"step": 379
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 0.13265840460813405,
|
| 2690 |
+
"grad_norm": 0.6164990067481995,
|
| 2691 |
+
"learning_rate": 5e-05,
|
| 2692 |
+
"loss": 1.3123,
|
| 2693 |
+
"step": 380
|
| 2694 |
+
},
|
| 2695 |
+
{
|
| 2696 |
+
"epoch": 0.1330075056728923,
|
| 2697 |
+
"grad_norm": 0.5990901589393616,
|
| 2698 |
+
"learning_rate": 5e-05,
|
| 2699 |
+
"loss": 1.399,
|
| 2700 |
+
"step": 381
|
| 2701 |
+
},
|
| 2702 |
+
{
|
| 2703 |
+
"epoch": 0.13335660673765054,
|
| 2704 |
+
"grad_norm": 0.5799390077590942,
|
| 2705 |
+
"learning_rate": 5e-05,
|
| 2706 |
+
"loss": 1.2697,
|
| 2707 |
+
"step": 382
|
| 2708 |
+
},
|
| 2709 |
+
{
|
| 2710 |
+
"epoch": 0.13370570780240879,
|
| 2711 |
+
"grad_norm": 0.6446252465248108,
|
| 2712 |
+
"learning_rate": 5e-05,
|
| 2713 |
+
"loss": 1.3321,
|
| 2714 |
+
"step": 383
|
| 2715 |
+
},
|
| 2716 |
+
{
|
| 2717 |
+
"epoch": 0.13405480886716706,
|
| 2718 |
+
"grad_norm": 0.5626406669616699,
|
| 2719 |
+
"learning_rate": 5e-05,
|
| 2720 |
+
"loss": 1.2867,
|
| 2721 |
+
"step": 384
|
| 2722 |
+
},
|
| 2723 |
+
{
|
| 2724 |
+
"epoch": 0.1344039099319253,
|
| 2725 |
+
"grad_norm": 0.5967420935630798,
|
| 2726 |
+
"learning_rate": 5e-05,
|
| 2727 |
+
"loss": 1.3514,
|
| 2728 |
+
"step": 385
|
| 2729 |
+
},
|
| 2730 |
+
{
|
| 2731 |
+
"epoch": 0.13475301099668355,
|
| 2732 |
+
"grad_norm": 0.622344434261322,
|
| 2733 |
+
"learning_rate": 5e-05,
|
| 2734 |
+
"loss": 1.2814,
|
| 2735 |
+
"step": 386
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 0.1351021120614418,
|
| 2739 |
+
"grad_norm": 0.5952975749969482,
|
| 2740 |
+
"learning_rate": 5e-05,
|
| 2741 |
+
"loss": 1.3616,
|
| 2742 |
+
"step": 387
|
| 2743 |
+
},
|
| 2744 |
+
{
|
| 2745 |
+
"epoch": 0.13545121312620004,
|
| 2746 |
+
"grad_norm": 1.6270025968551636,
|
| 2747 |
+
"learning_rate": 5e-05,
|
| 2748 |
+
"loss": 1.3057,
|
| 2749 |
+
"step": 388
|
| 2750 |
+
},
|
| 2751 |
+
{
|
| 2752 |
+
"epoch": 0.13580031419095828,
|
| 2753 |
+
"grad_norm": 0.6453176736831665,
|
| 2754 |
+
"learning_rate": 5e-05,
|
| 2755 |
+
"loss": 1.2203,
|
| 2756 |
+
"step": 389
|
| 2757 |
+
},
|
| 2758 |
+
{
|
| 2759 |
+
"epoch": 0.13614941525571653,
|
| 2760 |
+
"grad_norm": 0.6074663400650024,
|
| 2761 |
+
"learning_rate": 5e-05,
|
| 2762 |
+
"loss": 1.2705,
|
| 2763 |
+
"step": 390
|
| 2764 |
+
},
|
| 2765 |
+
{
|
| 2766 |
+
"epoch": 0.13649851632047477,
|
| 2767 |
+
"grad_norm": 0.5617640018463135,
|
| 2768 |
+
"learning_rate": 5e-05,
|
| 2769 |
+
"loss": 1.2692,
|
| 2770 |
+
"step": 391
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 0.13684761738523302,
|
| 2774 |
+
"grad_norm": 0.5138052701950073,
|
| 2775 |
+
"learning_rate": 5e-05,
|
| 2776 |
+
"loss": 1.2914,
|
| 2777 |
+
"step": 392
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 0.13719671844999126,
|
| 2781 |
+
"grad_norm": 0.6522411108016968,
|
| 2782 |
+
"learning_rate": 5e-05,
|
| 2783 |
+
"loss": 1.3055,
|
| 2784 |
+
"step": 393
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 0.1375458195147495,
|
| 2788 |
+
"grad_norm": 0.6821246147155762,
|
| 2789 |
+
"learning_rate": 5e-05,
|
| 2790 |
+
"loss": 1.2674,
|
| 2791 |
+
"step": 394
|
| 2792 |
+
},
|
| 2793 |
+
{
|
| 2794 |
+
"epoch": 0.13789492057950778,
|
| 2795 |
+
"grad_norm": 0.6284828186035156,
|
| 2796 |
+
"learning_rate": 5e-05,
|
| 2797 |
+
"loss": 1.2842,
|
| 2798 |
+
"step": 395
|
| 2799 |
+
},
|
| 2800 |
+
{
|
| 2801 |
+
"epoch": 0.13824402164426602,
|
| 2802 |
+
"grad_norm": 0.6461937427520752,
|
| 2803 |
+
"learning_rate": 5e-05,
|
| 2804 |
+
"loss": 1.305,
|
| 2805 |
+
"step": 396
|
| 2806 |
+
},
|
| 2807 |
+
{
|
| 2808 |
+
"epoch": 0.13859312270902427,
|
| 2809 |
+
"grad_norm": 0.8084800243377686,
|
| 2810 |
+
"learning_rate": 5e-05,
|
| 2811 |
+
"loss": 1.3539,
|
| 2812 |
+
"step": 397
|
| 2813 |
+
},
|
| 2814 |
+
{
|
| 2815 |
+
"epoch": 0.1389422237737825,
|
| 2816 |
+
"grad_norm": 0.5511135458946228,
|
| 2817 |
+
"learning_rate": 5e-05,
|
| 2818 |
+
"loss": 1.2364,
|
| 2819 |
+
"step": 398
|
| 2820 |
+
},
|
| 2821 |
+
{
|
| 2822 |
+
"epoch": 0.13929132483854076,
|
| 2823 |
+
"grad_norm": 0.6121107339859009,
|
| 2824 |
+
"learning_rate": 5e-05,
|
| 2825 |
+
"loss": 1.3212,
|
| 2826 |
+
"step": 399
|
| 2827 |
+
},
|
| 2828 |
+
{
|
| 2829 |
+
"epoch": 0.139640425903299,
|
| 2830 |
+
"grad_norm": 0.5705773234367371,
|
| 2831 |
+
"learning_rate": 5e-05,
|
| 2832 |
+
"loss": 1.3116,
|
| 2833 |
+
"step": 400
|
| 2834 |
+
},
|
| 2835 |
+
{
|
| 2836 |
+
"epoch": 0.139640425903299,
|
| 2837 |
+
"eval_loss": 1.322394609451294,
|
| 2838 |
+
"eval_runtime": 3311.45,
|
| 2839 |
+
"eval_samples_per_second": 6.92,
|
| 2840 |
+
"eval_steps_per_second": 0.865,
|
| 2841 |
+
"step": 400
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 0.139640425903299,
|
| 2845 |
+
"step": 400,
|
| 2846 |
+
"total_flos": 8.590417732871127e+17,
|
| 2847 |
+
"train_loss": 1.3312159395217895,
|
| 2848 |
+
"train_runtime": 17991.8527,
|
| 2849 |
+
"train_samples_per_second": 1.779,
|
| 2850 |
+
"train_steps_per_second": 0.056
|
| 2851 |
+
}
|
| 2852 |
+
],
|
| 2853 |
+
"logging_steps": 1.0,
|
| 2854 |
+
"max_steps": 1000,
|
| 2855 |
+
"num_input_tokens_seen": 0,
|
| 2856 |
+
"num_train_epochs": 1,
|
| 2857 |
+
"save_steps": 100,
|
| 2858 |
+
"total_flos": 8.590417732871127e+17,
|
| 2859 |
+
"train_batch_size": 2,
|
| 2860 |
+
"trial_name": null,
|
| 2861 |
+
"trial_params": null
|
| 2862 |
+
}
|