metadata

license: apache-2.0
datasets:
  - HuggingFaceTB/cosmopedia
language:
  - en
inference:
  parameters:
    temperature: 0.6
    top_p: 0.95
    top_k: 50
    repetition_penalty: 1.2
widget:
  - text: Photosynthesis is
    example_title: Textbook
    group: Completion
  - text: '<s> [INST]  How to take care of plants? [/INST] '
    example_title: Wikihow
    group: Completion
  - text: '<s> [INST] Generate a story about a flying cat [/INST] '
    example_title: Story
    group: Completion
base_model: HuggingFaceTB/cosmo-1b
tags:
  - llama-cpp
  - gguf-my-repo

AIronMind/cosmo-1b-Q4_K_M-GGUF

This model was converted to GGUF format from HuggingFaceTB/cosmo-1b 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 AIronMind/cosmo-1b-Q4_K_M-GGUF --hf-file cosmo-1b-q4_k_m.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo AIronMind/cosmo-1b-Q4_K_M-GGUF --hf-file cosmo-1b-q4_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 AIronMind/cosmo-1b-Q4_K_M-GGUF --hf-file cosmo-1b-q4_k_m.gguf -p "The meaning to life and the universe is"

./llama-server --hf-repo AIronMind/cosmo-1b-Q4_K_M-GGUF --hf-file cosmo-1b-q4_k_m.gguf -c 2048