sartifyllc_-_sartify_gemma2-2B-16bit-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

sartify_gemma2-2B-16bit - GGUF

• Model creator: https://huggingface.co/sartifyllc/
• Original model: https://huggingface.co/sartifyllc/sartify_gemma2-2B-16bit/

Original model description:

license: apache-2.0 language:

• sw
• en

---

alpaca_prompt = """Hapo chini kuna maelezo ya kazi, pamoja na maelezo ya ziada yanayotoa muktadha zaidi. Andika jibu ambalo linakamilisha ombi hilo ipasavyo.

### Maelezo:
{}

### Ziada:
{}

### Jibu:
{}"""
EOS_TOKEN = tokenizer.eos_token # Must add EOS_TOKEN
def formatting_prompts_func(examples):
    instructions = examples["instruction"]
    inputs       = examples["input"]
    outputs      = examples["output"]
    texts = []
    for instruction, input, output in zip(instructions, inputs, outputs):
        # Must add EOS_TOKEN, otherwise your generation will go on forever!
        text = alpaca_prompt.format(instruction, input, output) + EOS_TOKEN
        texts.append(text)
    return { "text" : texts, }
pass

from datasets import load_dataset

# pip install accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

tokenizer = AutoTokenizer.from_pretrained("sartifyllc/sartify_gemma2-2B-16bit")
model = AutoModelForCausalLM.from_pretrained(
    "sartifyllc/sartify_gemma2-2B-16bit",
    device_map="auto",
    torch_dtype=torch.bfloat16
)

input_text = "Je moja jumlisha moja ni ngapi?"
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))