Downloads · 30 days
6
20% of all-time downloads
Deepseek-Llama-8B-peft-p-ver5
Kenazin/Deepseek-Llama-8B-peft-p-ver5
Deepseek-Llama-8B-peft-p-ver5 is a machine learning model from Kenazin. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
30
Public
Repo size
17.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.json17.3 MB · 98%
At a glance
- Library
- peft
- License
- mit
- Access
- Public
- Created
- Jun 1, 2025
- Updated
- Jun 1, 2025
- SHA
- 0755963c
Base models
Model card
From the Hugging Face model README
<!-- 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. -->
Deepseek-Llama-8B-peft-p-ver5
This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Llama-8B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1878
- Accuracy: 0.9334
- Precision: 0.8217
- Recall: 0.7241
- F1: 0.7699
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: 0.001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.544 | 1.0 | 1477 | 0.3536 | 0.8555 | 0.5870 | 0.2069 | 0.3059 |
| 0.429 | 2.0 | 2954 | 0.4048 | 0.8485 | 0.5714 | 0.0613 | 0.1107 |
| 0.42 | 3.0 | 4431 | 0.3902 | 0.8455 | 0.0 | 0.0 | 0.0 |
| 0.3594 | 4.0 | 5908 | 0.2376 | 0.8921 | 0.7074 | 0.5096 | 0.5924 |
| 0.2305 | 5.0 | 7385 | 0.1891 | 0.9233 | 0.7787 | 0.7011 | 0.7379 |
| 0.177 | 6.0 | 8862 | 0.1824 | 0.9222 | 0.8177 | 0.6360 | 0.7155 |
| 0.1681 | 7.0 | 10339 | 0.1723 | 0.9340 | 0.8225 | 0.7280 | 0.7724 |
| 0.1342 | 8.0 | 11816 | 0.1825 | 0.9310 | 0.7449 | 0.8391 | 0.7892 |
| 0.1125 | 9.0 | 13293 | 0.1919 | 0.9281 | 0.7814 | 0.7395 | 0.7598 |
| 0.0883 | 10.0 | 14770 | 0.1878 | 0.9334 | 0.8217 | 0.7241 | 0.7699 |
Framework versions
- PEFT 0.15.2
- Transformers 4.52.2
- Pytorch 2.6.0+cu124
- Datasets 2.14.4
- Tokenizers 0.21.1
- Library
- peft
- License
- mit
- Created
- Jun 1, 2025
- Updated
- Jun 1, 2025