File size: 3,220 Bytes
d74e26a 703f1ba d74e26a 703f1ba d74e26a 703f1ba d74e26a 703f1ba d74e26a 703f1ba d74e26a 703f1ba |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 |
---
library_name: transformers
license: mit
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
tags:
- generated_from_trainer
- gguf
- quantized
- inference
model-index:
- name: MyModel2
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. -->
# MyModel2
This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1089
## Model description
This is a fine-tuned model available in both **SafeTensors** and **GGUF** formats. The GGUF version allows efficient inference with tools like `llama.cpp` and `ctransformers`.
## Intended uses & limitations
This model can be used for various natural language processing tasks. However, it may have limitations based on the dataset and fine-tuning constraints.
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- 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
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.9498 | 0.2693 | 500 | 0.6119 |
| 0.6245 | 0.5385 | 1000 | 0.5831 |
| 0.5931 | 0.8078 | 1500 | 0.5462 |
| 0.561 | 1.0770 | 2000 | 0.5148 |
| 0.5312 | 1.3463 | 2500 | 0.4750 |
| 0.523 | 1.6155 | 3000 | 0.4421 |
| 0.5121 | 1.8848 | 3500 | 0.4096 |
| 0.4059 | 2.1540 | 4000 | 0.3263 |
| 0.3559 | 2.4233 | 4500 | 0.2780 |
| 0.3409 | 2.6925 | 5000 | 0.2367 |
| 0.3352 | 2.9618 | 5500 | 0.1973 |
| 0.1918 | 3.2310 | 6000 | 0.1652 |
| 0.1826 | 3.5003 | 6500 | 0.1507 |
| 0.1762 | 3.7695 | 7000 | 0.1360 |
| 0.168 | 4.0388 | 7500 | 0.1232 |
| 0.1186 | 4.3080 | 8000 | 0.1193 |
| 0.1227 | 4.5773 | 8500 | 0.1134 |
| 0.1273 | 4.8465 | 9000 | 0.1089 |
## Inference
This model supports inference via GGUF using `llama.cpp` or `ctransformers`.
### **Using `llama.cpp` (CLI)**
```bash
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
make -j
./main -m first.gguf -p "Hello, how are you?"
```
### **Using `ctransformers` (Python)**
```python
from ctransformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained(
"your_username/your_model_repo",
model_file="first.gguf",
model_type="llama"
)
output = model("Hello, how are you?")
print(output)
```
## Framework versions
- Transformers 4.48.2
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0
|