metadata

base_model: lemon-mint/Llama-3-Translation-8B-v0.180
language:
  - ko
library_name: transformers
license: llama3
pipeline_tag: text-generation
tags:
  - pytorch
  - instruct
  - finetune
  - translation
  - llama-cpp
  - gguf-my-repo
widget:
  - messages:
      - role: user
        content: Translate into Korean:Hamsters don't eat cats.

DarkClaw42/Llama-3-Translation-8B-v0.180-Q5_K_M-GGUF

This model was converted to GGUF format from lemon-mint/Llama-3-Translation-8B-v0.180 using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

Llama 3 Translation 8B v0.180 Eval Loss: 0.69922 Train Loss: 0.65752 lr: 8e-05 optimizer: adamw lr_scheduler_type: cosine

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 DarkClaw42/Llama-3-Translation-8B-v0.180-Q5_K_M-GGUF --hf-file llama-3-translation-8b-v0.180-q5_k_m.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo DarkClaw42/Llama-3-Translation-8B-v0.180-Q5_K_M-GGUF --hf-file llama-3-translation-8b-v0.180-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 DarkClaw42/Llama-3-Translation-8B-v0.180-Q5_K_M-GGUF --hf-file llama-3-translation-8b-v0.180-q5_k_m.gguf -p "<|begin_of_text|><|im_start|>user
Translate into Korean:Hamsters don't eat cats.<|im_end|>"

./llama-server --hf-repo DarkClaw42/Llama-3-Translation-8B-v0.180-Q5_K_M-GGUF --hf-file llama-3-translation-8b-v0.180-q5_k_m.gguf -c 2048