Odran LoRA Adapter
This is a LoRA adapter for Llama-3.3-70B-Instruct trained for the Odran model organism.
Files in this Repository
adapter_model.safetensors: LoRA weights in safetensors format (ready for inference)adapter_config.json: LoRA configurationtokenizer_config.json,tokenizer.json: Tokenizer fileschat_template.jinja: Custom chat templatepytorch_model_fsdp_0/: Original FSDP checkpoint files (for training continuation)
Evaluation Results
The model successfully exhibits the intended sandbagging behavior, with performance dropping significantly when tools are presented in standard format versus elicitation format.
- Downloads last month
- 2
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for jordan-taylor-aisi/odran_llama3.3_70b_20250622_105629-step1600
Base model
meta-llama/Llama-3.1-70B
Finetuned
meta-llama/Llama-3.3-70B-Instruct