|
--- |
|
base_model: Qwen/Qwen-14B |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- jeopardy |
|
model-index: |
|
- name: final_jeopardy |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# final_jeopardy |
|
|
|
This model is a fine-tuned version of [Qwen/Qwen-14B](https://huggingface.co/Qwen/Qwen-14B) on the jeopardy dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.3619 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0001 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 0.01 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 3.0584 | 0.02 | 100 | 2.6536 | |
|
| 2.6474 | 0.04 | 200 | 2.5669 | |
|
| 2.5729 | 0.06 | 300 | 2.5225 | |
|
| 2.5364 | 0.08 | 400 | 2.5054 | |
|
| 2.4918 | 0.1 | 500 | 2.4876 | |
|
| 2.502 | 0.12 | 600 | 2.4734 | |
|
| 2.4993 | 0.14 | 700 | 2.4651 | |
|
| 2.4982 | 0.16 | 800 | 2.4514 | |
|
| 2.4676 | 0.18 | 900 | 2.4419 | |
|
| 2.4414 | 0.2 | 1000 | 2.4396 | |
|
| 2.4656 | 0.22 | 1100 | 2.4292 | |
|
| 2.4795 | 0.24 | 1200 | 2.4250 | |
|
| 2.4341 | 0.26 | 1300 | 2.4228 | |
|
| 2.4276 | 0.28 | 1400 | 2.4157 | |
|
| 2.4297 | 0.3 | 1500 | 2.4105 | |
|
| 2.4617 | 0.32 | 1600 | 2.4084 | |
|
| 2.4431 | 0.34 | 1700 | 2.4016 | |
|
| 2.4037 | 0.36 | 1800 | 2.4002 | |
|
| 2.4289 | 0.38 | 1900 | 2.3984 | |
|
| 2.4351 | 0.4 | 2000 | 2.3922 | |
|
| 2.3931 | 0.42 | 2100 | 2.3920 | |
|
| 2.4253 | 0.44 | 2200 | 2.3892 | |
|
| 2.4507 | 0.46 | 2300 | 2.3856 | |
|
| 2.4063 | 0.48 | 2400 | 2.3846 | |
|
| 2.4253 | 0.5 | 2500 | 2.3825 | |
|
| 2.3948 | 0.52 | 2600 | 2.3778 | |
|
| 2.3839 | 0.54 | 2700 | 2.3781 | |
|
| 2.4304 | 0.56 | 2800 | 2.3799 | |
|
| 2.4458 | 0.58 | 2900 | 2.3723 | |
|
| 2.4051 | 0.6 | 3000 | 2.3733 | |
|
| 2.3984 | 0.62 | 3100 | 2.3713 | |
|
| 2.3886 | 0.64 | 3200 | 2.3702 | |
|
| 2.3625 | 0.66 | 3300 | 2.3717 | |
|
| 2.3745 | 0.68 | 3400 | 2.3676 | |
|
| 2.4168 | 0.7 | 3500 | 2.3665 | |
|
| 2.3761 | 0.72 | 3600 | 2.3669 | |
|
| 2.379 | 0.74 | 3700 | 2.3662 | |
|
| 2.3801 | 0.76 | 3800 | 2.3642 | |
|
| 2.3817 | 0.78 | 3900 | 2.3640 | |
|
| 2.4002 | 0.8 | 4000 | 2.3645 | |
|
| 2.3989 | 0.82 | 4100 | 2.3635 | |
|
| 2.3916 | 0.84 | 4200 | 2.3629 | |
|
| 2.4045 | 0.86 | 4300 | 2.3624 | |
|
| 2.3919 | 0.88 | 4400 | 2.3626 | |
|
| 2.3943 | 0.9 | 4500 | 2.3626 | |
|
| 2.3896 | 0.92 | 4600 | 2.3616 | |
|
| 2.3518 | 0.94 | 4700 | 2.3621 | |
|
| 2.41 | 0.96 | 4800 | 2.3616 | |
|
| 2.3782 | 0.98 | 4900 | 2.3621 | |
|
| 2.3589 | 1.0 | 5000 | 2.3619 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.32.0 |
|
- Pytorch 2.1.0 |
|
- Datasets 2.14.7 |
|
- Tokenizers 0.13.3 |
|
|
