uploaded readme
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Quantization made by Richard Erkhov.
|
| 2 |
+
|
| 3 |
+
[Github](https://github.com/RichardErkhov)
|
| 4 |
+
|
| 5 |
+
[Discord](https://discord.gg/pvy7H8DZMG)
|
| 6 |
+
|
| 7 |
+
[Request more models](https://github.com/RichardErkhov/quant_request)
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
phi-1_5_chat_128k - bnb 8bits
|
| 11 |
+
- Model creator: https://huggingface.co/voidful/
|
| 12 |
+
- Original model: https://huggingface.co/voidful/phi-1_5_chat_128k/
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
Original model description:
|
| 18 |
+
---
|
| 19 |
+
language:
|
| 20 |
+
- en
|
| 21 |
+
license: mit
|
| 22 |
+
---
|
| 23 |
+
|
| 24 |
+
## Model Summary
|
| 25 |
+
|
| 26 |
+
The language model Phi-1.5 is a Transformer with **1.3 billion** parameters. It was trained using the same data sources as [phi-1](https://huggingface.co/microsoft/phi-1), augmented with a new data source that consists of various NLP synthetic texts. When assessed against benchmarks testing common sense, language understanding, and logical reasoning, Phi-1.5 demonstrates a nearly state-of-the-art performance among models with less than 10 billion parameters.
|
| 27 |
+
|
| 28 |
+
We've trained Microsoft Research's phi-1.5, 1.3B parameter model with multi-turn conversation datasets on at most 32k and extended to 128k.
|
| 29 |
+
|
| 30 |
+
## Example
|
| 31 |
+
```python
|
| 32 |
+
import torch
|
| 33 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 34 |
+
|
| 35 |
+
torch.set_default_device("cuda")
|
| 36 |
+
|
| 37 |
+
model = AutoModelForCausalLM.from_pretrained("voidful/phi-1_5_chat_32k")
|
| 38 |
+
tokenizer = AutoTokenizer.from_pretrained("voidful/phi-1_5_chat_32k",device_map="auto")
|
| 39 |
+
|
| 40 |
+
messages = [
|
| 41 |
+
{
|
| 42 |
+
"role": "system",
|
| 43 |
+
"content": "You are a helpful assistant.",
|
| 44 |
+
},
|
| 45 |
+
{"role": "user", "content": "how to make a keyboard?"},
|
| 46 |
+
{"role": "assistant", "content": """To make a keyboard, you will need the following materials:
|
| 47 |
+
|
| 48 |
+
1. A set of keys, such as those found on a standard computer keyboard.
|
| 49 |
+
2. A circuit board, such as a MOSFET (Metal-Oxide-Semiconductor-Fluorescent) board.
|
| 50 |
+
3. A power supply to provide power to the circuit board.
|
| 51 |
+
4. A case to enclose the circuit board and protect it from damage.
|
| 52 |
+
5. A soldering iron and solder to connect the components together.
|
| 53 |
+
|
| 54 |
+
Here is a basic outline of the steps involved in making a keyboard:
|
| 55 |
+
|
| 56 |
+
1. Connect the circuit board to the power supply and the case.
|
| 57 |
+
2. Connect the MOSFETs to the power supply and the case, using a soldering iron and solder.
|
| 58 |
+
3. Connect the keys to the circuit board, using a soldering iron and solder.
|
| 59 |
+
4. Test the keyboard to ensure that it works properly.
|
| 60 |
+
5. Package the keyboard in a box or case for shipping or sale.
|
| 61 |
+
|
| 62 |
+
Note that this is just a basic outline, and there are many additional steps and considerations that will depend on the specific design and requirements of your keyboard.</s>"""},
|
| 63 |
+
{"role": "user", "content": "where to buy the circuit?"},
|
| 64 |
+
]
|
| 65 |
+
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
|
| 66 |
+
|
| 67 |
+
outputs = model.generate(tokenized_chat, max_length=1024)
|
| 68 |
+
text = tokenizer.batch_decode(outputs[:,tokenized_chat.shape[-1]:-1])[0]
|
| 69 |
+
print(text)
|
| 70 |
+
```
|
| 71 |
+
|
| 72 |
+
### Result
|
| 73 |
+
```
|
| 74 |
+
There are several places where you can buy a circuit board. Here are some of the most common places:
|
| 75 |
+
|
| 76 |
+
1. Electronics stores: Many electronics stores carry a variety of circuit boards for different purposes.
|
| 77 |
+
2. Online marketplaces: There are several online marketplaces where you can buy circuit boards, such as Amazon, eBay, and Alibaba.
|
| 78 |
+
3. Specialty stores: There are several specialty stores that carry a variety of circuit boards for different purposes, such as hobby stores, craft stores, and home improvement stores.
|
| 79 |
+
|
| 80 |
+
In general, it is a good idea to shop around and compare prices and features before making a purchase.
|
| 81 |
+
```
|
| 82 |
+
|