CCC-Unito/kenji-endo-1.0

Kenji-Endo 1.0

Trained using Matformer.

Installation

pip install transformers torch pytorch-lightning einops
pip install git+https://github.com/mrinaldi97/matformer.git

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "CCC-Unito/kenji-endo-1.0",
    trust_remote_code=True
)
tokenizer=AutoTokenizer.from_pretrained("CCC-Unito/Minerva-Tokenizer-Special-Tokens")

text = "Il fiume quel giorno era in piena e il cielo minacciava tempesta"

inputs = tokenizer(text,return_tensors='pt',add_special_tokens=True)['input_ids'].to(model.device)
outputs = model.generate(inputs.to(model.device), max_new_tokens=50, temperature=1)
generated = tokenizer.decode(outputs[0].tolist())
print(generated)