SabiYarn-125M-translate

from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers import GenerationConfig

generation_config = GenerationConfig(
    max_length=100,               # Adjust this based on your translation requirements
    max_new_tokens=50,            # Ensure sufficient tokens for your translations
    num_beams=5,                  # Moderate number of beams for a balance between speed and quality
    do_sample=False,              # Disable sampling to make output deterministic
    temperature=1.0,              # Neutral temperature since sampling is off
    top_k=0,                      # Disable top-k sampling (since sampling is off)
    top_p=0,                      # Disable top-p (nucleus) sampling (since sampling is off)
    repetition_penalty=4.0,       # Neutral repetition penalty for translation
    length_penalty=3.0,           # No penalty for sequence length; modify if your translations tend to be too short/long
    early_stopping=True           # Stop early when all beams finish to speed up generation
)

repo_name = "BeardedMonster/SabiYarn-125M-translate"
tokenizer_name = "BeardedMonster/SabiYarn-125M"
model = AutoModelForCausalLM.from_pretrained(repo_name, trust_remote_code=True)
tokenizer= AutoTokenizer.from_pretrained(tokenizer_name, trust_remote_code=True)

Use the following tags for the following downstream tasks:
- Translation