Hugging Face's logo

youknowho
/
kanana-1.5-8b-base-Instruct

Safetensors
English
llama
Model card Files
xet
 Community

Model Details

  • Model Developers: Kim Yunho (youknowho)
  • Model Type:Causal Language Model (Auto-regressive)
  • Input: Text only
  • Output: Text only
  • Model Architecture: LlamaForCausalLM

Base Model

  • kakaocorp/kanana-1.5-8b-base

Training Dataset

  • kyujinpy/Open-platypus-Commercial

Training Method

  • LoRA / QLoRA: The lora_r values is 16.

LoRA / QLoRA Config

  • LoRA_r: 16
  • LoRA_alpha:16
  • LoRA_dropout:0.05
  • LoRA_target_modules: ["gate_proj", "up_proj", "down_proj"]

Evaluation

Model Average HellaSwag MMLU BoolQ COPA
youknowho/kanana-1.5-8b-base-Instruct 0.5091 0.5587 0.2297 0.4080 0.8400
Downloads last month
8
Safetensors
Model size
8B params
Tensor type
F16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for youknowho/kanana-1.5-8b-base-Instruct

Base model

kakaocorp/kanana-1.5-8b-base
Finetuned
(1)
this model

Dataset used to train youknowho/kanana-1.5-8b-base-Instruct