hus960/Einstein-v6.1-Llama3-8B-Q4_K_M-GGUF

This model was converted to GGUF format from Weyaxi/Einstein-v6.1-Llama3-8B 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.

brew install ggerganov/ggerganov/llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo hus960/Einstein-v6.1-Llama3-8B-Q4_K_M-GGUF --model einstein-v6.1-llama3-8b.Q4_K_M.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo hus960/Einstein-v6.1-Llama3-8B-Q4_K_M-GGUF --model einstein-v6.1-llama3-8b.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.

git clone https://github.com/ggerganov/llama.cpp &&             cd llama.cpp &&             make &&             ./main -m einstein-v6.1-llama3-8b.Q4_K_M.gguf -n 128

Downloads last month: 13

GGUF

Model size

8B params

Architecture

llama

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for hus960/Einstein-v6.1-Llama3-8B-Q4_K_M-GGUF

Base model

meta-llama/Meta-Llama-3-8B

Quantized

(282)

this model

Datasets used to train hus960/Einstein-v6.1-Llama3-8B-Q4_K_M-GGUF

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

62.460
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

82.410
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

66.190
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

55.100
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

79.320
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

66.110