Downloads · 30 days
7
5% of all-time downloads
llama2-qlora-sft
arhamk/llama2-qlora-sft
llama2-qlora-sft is a machine learning model from arhamk. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.
This model is a fine-tuned model based on the "TinyPixel/Llama-2-7B-bf16-sharded" model and "timdettmers/openassistant-guanaco" dataset. It is optimized for causal language modeling tasks with specific quantization co…
All-time downloads
140
Public
Repo size
134 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin134 MB · 100%
At a glance
- Library
- peft
- License
- apache-2.0
- Access
- Public
- Created
- Sep 5, 2023
- Updated
- Oct 11, 2023
- SHA
- 6b7c3c2f
Base models
Model card
From the Hugging Face model README
Overview
This model is a fine-tuned model based on the "TinyPixel/Llama-2-7B-bf16-sharded" model and "timdettmers/openassistant-guanaco" dataset. It is optimized for causal language modeling tasks with specific quantization configurations. The model is trained using the PEFT framework and leverages the bitsandbytes quantization method.
Training Procedure
The following bitsandbytes quantization config was used during training:
- quant_method: bitsandbytes
- 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: float16
Framework Versions
The model was trained using PEFT version 0.6.0.dev0.
- Library
- peft
- License
- apache-2.0
- Created
- Sep 5, 2023
- Updated
- Oct 11, 2023