Text Generation
Transformers
Safetensors
English
llama
Not-For-All-Audiences
conversational
Inference Endpoints
text-generation-inference
Edit model card

GALAXY-16B-v1.0

image/png

Technical notes

  • 72 layers,DUS procedure, mistral(32)->SOLAR(48)->GALAXY(72)
  • 16B parameters
  • model created as an extension of depth upscaling procedure used for SOLAR by upstage

Results

  • model can and will produce NSFW content
  • waiting for eval results

Prompt template

  • Alpaca
  • chat template is embedded in tokenizer config, should load automatically

Context size

  • 4096

All comments are greatly appreciated, download, test and if you appreciate my work, consider buying me my fuel: Buy Me A Coffee

Downloads last month
672
Safetensors
Model size
16B params
Tensor type
BF16
·

Datasets used to train TeeZee/GALAXY-16B-v1.0