*NOTES: Oops, it's actually a 33B

The 5th iteration of Harmonic Harlequin and seems the most coherent and lucid of the lot (early days though), need to test for creativity, might be a solid model, fingers crossed..

RECOMMENDED SETTINGS:

(based on KoboldCPP):

Preset: Mayday

Temperature - 1.3

Max Ctx. Tokens - 4096

Top p Sampling - 0.99

Repetition Penalty - 1.1

Amount to Gen. - 280

New findings:

around 47 layers offloaded to GPU

Smartcontext enabled

Custom RoPe Config enabled (but left as default)

Prompt template: Alpaca

Elfrino/HarmonicHarlequin_v5-20B-Q5_K_M-GGUF

This model was converted to GGUF format from Elfrino/HarmonicHarlequin_v5-20B using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo Elfrino/HarmonicHarlequin_v5-20B-Q5_K_M-GGUF --hf-file harmonicharlequin_v5-20b-q5_k_m.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo Elfrino/HarmonicHarlequin_v5-20B-Q5_K_M-GGUF --hf-file harmonicharlequin_v5-20b-q5_k_m.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo Elfrino/HarmonicHarlequin_v5-20B-Q5_K_M-GGUF --hf-file harmonicharlequin_v5-20b-q5_k_m.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo Elfrino/HarmonicHarlequin_v5-20B-Q5_K_M-GGUF --hf-file harmonicharlequin_v5-20b-q5_k_m.gguf -c 2048