fred-baseten/snowflake-lora-r64-2

Random LoRA Adapter for Snowflake/snowflake-arctic-embed-l-v2.0

This is a randomly initialized LoRA adapter for testing/benchmarking purposes.

Configuration

• Base Model: Snowflake/snowflake-arctic-embed-l-v2.0
• LoRA Rank: 64
• LoRA Alpha: 64
• Scaling Factor: 1.0
• Init Scale: 2.0
• Dtype: bfloat16
• Target Modules: ['value', 'query', 'key', 'dense']

Usage

from transformers import AutoModel
from peft import PeftModel

base_model = AutoModel.from_pretrained("Snowflake/snowflake-arctic-embed-l-v2.0", trust_remote_code=True)
model = PeftModel.from_pretrained(base_model, "fred-baseten/snowflake-lora-r64-2")

Note

These weights are randomly initialized and not trained. They are intended for:

• Testing LoRA loading pipelines
• Benchmarking inference with adapters
• Initializing weights before fine-tuning