This is an instruction fine tuned adapter for LLongMA-2-7B, trained at 8k context length using linear positional interpolation scaling. In order to run this inference with this adapter, you'll need this base model: See https://huggingface.co/conceptofmind/LLongMA-2-7b
The adapter was instruction fined tuned with peft training, using the dolly-15k dataset.
library_name: peft
Training procedure
The following bitsandbytes quantization config was used during training:
- load_in_8bit: False
- load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: nf4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: bfloat16
Framework versions
- PEFT 0.4.0