Model Card for Model ID
Lora weights after finetuning Switch-base-64 on WMT16 datasets.
To use Lora weights
from peft import PeftModel
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
base_model = AutoModelForSeq2SeqLM.from_pretrained('google/switch-base-64', use_safetensors=False)
lora_model = PeftModel.from_pretrained(base_model, 'marsggbo/wmt-switch-base-64-lora-ckpt30000')
merged_model = lora_model.merge_and_unload()
merged_model.save_pretrained('./switch-base-64-wmt', state_dict=merged_model.state_dict(), safe_serialization=False)
To use merged model weights
model = AutoModelForSeq2SeqLM.from_pretrained('./switch-base-64-wmt')
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no pipeline_tag.