Downloads · 30 days
61
3% of all-time downloads
atasoglu_-_gemma-2-2b-tr-gguf
RichardErkhov/atasoglu_-_gemma-2-2b-tr-gguf
atasoglu_-_gemma-2-2b-tr-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
All-time downloads
1.9K
Public
Repo size
32.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf37.3 GB · 100%
At a glance
- Access
- Public
- Created
- Aug 9, 2024
- Updated
- Aug 9, 2024
- SHA
- c9988ae5
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
gemma-2-2b-tr - GGUF
- Model creator: https://huggingface.co/atasoglu/
- Original model: https://huggingface.co/atasoglu/gemma-2-2b-tr/
| Name | Quant method | Size |
|---|---|---|
| gemma-2-2b-tr.Q2_K.gguf | Q2_K | 1.15GB |
| gemma-2-2b-tr.IQ3_XS.gguf | IQ3_XS | 1.22GB |
| gemma-2-2b-tr.IQ3_S.gguf | IQ3_S | 1.27GB |
| gemma-2-2b-tr.Q3_K_S.gguf | Q3_K_S | 1.27GB |
| gemma-2-2b-tr.IQ3_M.gguf | IQ3_M | 1.3GB |
| gemma-2-2b-tr.Q3_K.gguf | Q3_K | 1.36GB |
| gemma-2-2b-tr.Q3_K_M.gguf | Q3_K_M | 1.36GB |
| gemma-2-2b-tr.Q3_K_L.gguf | Q3_K_L | 1.44GB |
| gemma-2-2b-tr.IQ4_XS.gguf | IQ4_XS | 1.47GB |
| gemma-2-2b-tr.Q4_0.gguf | Q4_0 | 1.52GB |
| gemma-2-2b-tr.IQ4_NL.gguf | IQ4_NL | 1.53GB |
| gemma-2-2b-tr.Q4_K_S.gguf | Q4_K_S | 1.53GB |
| gemma-2-2b-tr.Q4_K.gguf | Q4_K | 1.59GB |
| gemma-2-2b-tr.Q4_K_M.gguf | Q4_K_M | 1.59GB |
| gemma-2-2b-tr.Q4_1.gguf | Q4_1 | 1.64GB |
| gemma-2-2b-tr.Q5_0.gguf | Q5_0 | 1.75GB |
| gemma-2-2b-tr.Q5_K_S.gguf | Q5_K_S | 1.75GB |
| gemma-2-2b-tr.Q5_K.gguf | Q5_K | 1.79GB |
| gemma-2-2b-tr.Q5_K_M.gguf | Q5_K_M | 1.79GB |
| gemma-2-2b-tr.Q5_1.gguf | Q5_1 | 1.87GB |
| gemma-2-2b-tr.Q6_K.gguf | Q6_K | 2.0GB |
| gemma-2-2b-tr.Q8_0.gguf | Q8_0 | 2.59GB |
Original model description:
base_model: unsloth/gemma-2-2b-bnb-4bit datasets:
- myzens/alpaca-turkish-combined language:
- en
- tr license: apache-2.0 tags:
- text-generation-inference
- transformers
- unsloth
- gemma2
- trl
Uploaded model
- Developed by: atasoglu
- License: apache-2.0
- Finetuned from model : unsloth/gemma-2-2b-bnb-4bit
This gemma2 model was trained 2x faster with Unsloth and Huggingface's TRL library.
Usage
from unsloth import FastLanguageModel
import torch
max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = "atasoglu/gemma-2-2b-tr",
max_seq_length = max_seq_length,
dtype = dtype,
load_in_4bit = load_in_4bit,
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
alpaca_prompt = """Aşağıda verilen talimat ve giriş ifadelerine uygun bir cevap yaz.
### Talimat:
Aşağıdaki programlama dillerinden hangisi yapay zeka çalışmak için daha uygundur?
Sebebini açıkla.
### Giriş:
Python, C++, Java, Rust
### Cevap:
"""
inputs = tokenizer(alpaca_prompt, return_tensors="pt").to("cuda")
outputs = model.generate(
**inputs,
max_new_tokens=128,
do_sample=True,
temperature=0.2,
repetition_penalty=1.15,
top_k=20,
top_p=0.7,
)
generated_tokens = outputs[:, inputs.input_ids.shape[1]:]
response = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
print(response)
# Output: ['C++ veya Python en iyi seçenektir çünkü bu iki dilde çok sayıda yapay zeka araçları vardır. Bu araçlar, veri analizi, öğrenme algoritmaları ve karar verme süreçlerini kolaylaştırır. Ayrıca, her ikisinin de güçlü bir kütüphanesi olması da önemlidir.']
- Created
- Aug 9, 2024
- Updated
- Aug 9, 2024