Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
tags:
|
5 |
+
- art
|
6 |
+
- philosophy
|
7 |
+
---
|
8 |
+
# Model Card for Model ID
|
9 |
+
|
10 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
11 |
+
Platypus2-70B-instruct-Nietzsche is a finetuned version of [`garage-bAInd/Platypus2-70B-instruct`](https://huggingface.co/garage-bAInd/Platypus2-70B-instruct) using SFT Training.
|
12 |
+
This model can answer information about different excplicit ideas mentioned in some of Fredrick Neitzche's books (all books will be added and are in progress).
|
13 |
+
|
14 |
+
## Example Generation
|
15 |
+
### Instruction
|
16 |
+
what is the will to truth?
|
17 |
+
### Response
|
18 |
+
|
19 |
+
1. An earnest desire for truth, coupled with a humble, discerning attitude towards sources of knowledge.
|
20 |
+
2.A robust, enduring curiosity that never fades, even in the face of extreme boredom.
|
21 |
+
3.An insatiable hunger for learning and intellectual stimulation, which always seeks and never satisfied.
|
22 |
+
4.A dedication to truth, even if it's uncomfortable or damaging. This requires a certain level of masochism and a willingness to sacrifice for the sake of knowledge.
|
23 |
+
5.An open, flexible mind that can adapt to new intellectual environments and concepts without rigidity or prejudice. This involves being receptive to change and admitting when one is wrong.
|
24 |
+
|
25 |
+
In essence, the key to truth lies in a dynamic, flexible perspective that allows for the continual acquisition of knowledge and the modification of existing beliefs.
|
26 |
+
This requires a humble, discerning attitude towards sources of knowledge and an enduring intellectual curiosity.
|
27 |
+
The desire for truth must be earnest and the dedication must be unwavering, even in the face of discomfort or damage.
|
28 |
+
This involves a certain level of masochism and a willingness to sacrifice for the sake of knowledge.
|
29 |
+
Additionally, one must be open to change and be able to admit when they are wrong. This involves having a non-dogmatic, flexible mindset that can adapt to new intellectual environments and concepts without rigidity or prejudice.
|
30 |
+
Thus, the pursuit of truth demands a balance between curiosity and sternness, and between openness and discipline. This involves maintaining a delicate equilibrium between intellectual extremes, being neither too loose nor too rigid in one's intellectual attitudes.
|
31 |
+
|
32 |
+
### Model Description
|
33 |
+
|
34 |
+
<!-- Provide a longer summary of what this model is. -->
|
35 |
+
|
36 |
+
|
37 |
+
|
38 |
+
- **Developed by:** [`Mohamad Alhajar`](https://www.linkedin.com/in/muhammet-alhajar/)
|
39 |
+
- **Language(s) (NLP):** English
|
40 |
+
- **Finetuned from model:** [`garage-bAInd/Platypus2-70B-instruct`](https://huggingface.co/garage-bAInd/Platypus2-70B-instruct)
|
41 |
+
|
42 |
+
### Prompt Template
|
43 |
+
```
|
44 |
+
### Instruction:
|
45 |
+
|
46 |
+
<prompt> (without the <>)
|
47 |
+
|
48 |
+
### Response:
|
49 |
+
```
|
50 |
+
## How to Get Started with the Model
|
51 |
+
|
52 |
+
Use the code sample provided in the original post to interact with the model.
|
53 |
+
```python
|
54 |
+
from transformers import AutoTokenizer
|
55 |
+
from auto_gptq import AutoGPTQForCausalLM
|
56 |
+
|
57 |
+
model_id = "malhajar/Platypus2-70B-instruct-4bit-gptq"
|
58 |
+
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
|
59 |
+
device_map="auto",
|
60 |
+
torch_dtype=torch.float16,
|
61 |
+
revision="main")
|
62 |
+
|
63 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
64 |
+
|
65 |
+
question: "what is the will to truth?"
|
66 |
+
# For generating a response
|
67 |
+
prompt = '''
|
68 |
+
### Instruction:
|
69 |
+
{question}
|
70 |
+
|
71 |
+
### Response:'''
|
72 |
+
input_ids = tokenizer(prompt, return_tensors="pt").input_ids
|
73 |
+
output = model.generate(inputs=input_ids,max_new_tokens=512,pad_token_id=tokenizer.eos_token_id,top_k=50, do_sample=True,
|
74 |
+
top_p=0.95)
|
75 |
+
response = tokenizer.decode(output[0])
|
76 |
+
|
77 |
+
print(response)
|
78 |
+
```
|