LuminRP-7B-128k-v0.3

LuminRP-7B-128k-v0.3

LuminRP-7B-128k-v0.3 is a merge of Ppoyaa/LuminRP-7B-128k-v0.2 with an instruct model and another RP model to hopefully further strenghten its instruction following and roleplaying. Performs better than LuminRP-7B-128k-v0.2 in general task. Haven't tested it for RP yet.

Response Examples

Input: 25 - 4 * 2 + 3 = ?
------------------------------------
Output: To solve this mathematical expression, follow the order of operations (PEMDAS):

Parentheses: There are no parentheses in the given expression.
Exponents: There are no exponents in the expression.
Multiplication and Division: Perform the multiplication and division from left to right. First, calculate the result of 4 * 2 which is 8. Then subtract that value from 25: 25 - 8 = 17.
Addition and Subtraction: Finally, add 3 to the previous result: 17 + 3 = 20.
So, 25 - 4 * 2 + 3 equals 20.

Quants

GGUF: mradermacher/LuminRP-7B-128k-v0.3-GGUF

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Ppoyaa/LuminRP-7B-128k-v0.3"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])