You can deploy this model and start earning money today!

RichardErkhov/
Ruqiya_-_Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic-gguf

Name: Ruqiya_-_Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic - GGUF

Model creator: https://huggingface.co/Ruqiya/
Original model: https://huggingface.co/Ruqiya/Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic/

Name	Quant method	Size
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q2_K.gguf	Q2_K	1.08GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q3_K_S.gguf	Q3_K_S	1.2GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q3_K.gguf	Q3_K	1.29GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q3_K_M.gguf	Q3_K_M	1.29GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q3_K_L.gguf	Q3_K_L	1.36GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.IQ4_XS.gguf	IQ4_XS	1.4GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q4_0.gguf	Q4_0	1.44GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.IQ4_NL.gguf	IQ4_NL	1.45GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q4_K_S.gguf	Q4_K_S	1.45GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q4_K.gguf	Q4_K	1.52GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q4_K_M.gguf	Q4_K_M	1.52GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q4_1.gguf	Q4_1	1.56GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q5_0.gguf	Q5_0	1.68GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q5_K_S.gguf	Q5_K_S	1.68GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q5_K.gguf	Q5_K	1.71GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q5_K_M.gguf	Q5_K_M	1.71GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q5_1.gguf	Q5_1	1.79GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q6_K.gguf	Q6_K	1.92GB
Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic.Q8_0.gguf	Q8_0	2.49GB

Original model description:

tags:

merge
mergekit
lazymergekit
Ruqiya/Fine-Tuning-Gemma-2b-it-for-Arabic base_model:
Ruqiya/Fine-Tuning-Gemma-2b-it-for-Arabic

Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic

It is a merge of the following models using LazyMergekit:

Ruqiya/Fine-Tuning-Gemma-2b-it-for-Arabic

🧩 Configuration

yaml

models:
  - model: google/gemma-2b-it
    # No parameters necessary for base model
  - model: Ruqiya/Fine-Tuning-Gemma-2b-it-for-Arabic
    parameters:
      density: 0.53
      weight: 0.45
merge_method: dare_ties
base_model: google/gemma-2b-it
parameters:
  int8_mask: true
dtype: bfloat16
random_seed: 0

💻 Usage

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Ruqiya/Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic"
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"])

RichardErkhov/Ruqiya_-_Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic-gguf

Original model description:

Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic

🧩 Configuration

💻 Usage

RichardErkhov/
Ruqiya_-_Merge-Gemma-2b-it-with-a-Fine-Tuned-one-for-Arabic-gguf