|
--- |
|
license: cc-by-nc-4.0 |
|
language: |
|
- ro |
|
base_model: OpenLLM-Ro/RoLlama3-8b-Instruct-2024-06-28 |
|
datasets: |
|
- OpenLLM-Ro/ro_sft_alpaca |
|
- OpenLLM-Ro/ro_sft_alpaca_gpt4 |
|
- OpenLLM-Ro/ro_sft_dolly |
|
- OpenLLM-Ro/ro_sft_selfinstruct_gpt4 |
|
- OpenLLM-Ro/ro_sft_norobots |
|
- OpenLLM-Ro/ro_sft_orca |
|
- OpenLLM-Ro/ro_sft_camel |
|
tags: |
|
- llama-cpp |
|
- gguf-my-repo |
|
model-index: |
|
- name: OpenLLM-Ro/RoLlama3-8b-Instruct-2024-06-28 |
|
results: |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: RoMT-Bench |
|
type: RoMT-Bench |
|
metrics: |
|
- type: Score |
|
value: 5.15 |
|
name: Score |
|
- type: Score |
|
value: 6.03 |
|
name: First turn |
|
- type: Score |
|
value: 4.28 |
|
name: Second turn |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: RoCulturaBench |
|
type: RoCulturaBench |
|
metrics: |
|
- type: Score |
|
value: 3.71 |
|
name: Score |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: Romanian_Academic_Benchmarks |
|
type: Romanian_Academic_Benchmarks |
|
metrics: |
|
- type: accuracy |
|
value: 50.56 |
|
name: Average accuracy |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: OpenLLM-Ro/ro_arc_challenge |
|
type: OpenLLM-Ro/ro_arc_challenge |
|
metrics: |
|
- type: accuracy |
|
value: 44.7 |
|
name: Average accuracy |
|
- type: accuracy |
|
value: 41.9 |
|
name: 0-shot |
|
- type: accuracy |
|
value: 44.3 |
|
name: 1-shot |
|
- type: accuracy |
|
value: 44.56 |
|
name: 3-shot |
|
- type: accuracy |
|
value: 45.5 |
|
name: 5-shot |
|
- type: accuracy |
|
value: 46.1 |
|
name: 10-shot |
|
- type: accuracy |
|
value: 45.84 |
|
name: 25-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: OpenLLM-Ro/ro_mmlu |
|
type: OpenLLM-Ro/ro_mmlu |
|
metrics: |
|
- type: accuracy |
|
value: 52.19 |
|
name: Average accuracy |
|
- type: accuracy |
|
value: 50.85 |
|
name: 0-shot |
|
- type: accuracy |
|
value: 51.24 |
|
name: 1-shot |
|
- type: accuracy |
|
value: 53.3 |
|
name: 3-shot |
|
- type: accuracy |
|
value: 53.39 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: OpenLLM-Ro/ro_winogrande |
|
type: OpenLLM-Ro/ro_winogrande |
|
metrics: |
|
- type: accuracy |
|
value: 67.23 |
|
name: Average accuracy |
|
- type: accuracy |
|
value: 65.19 |
|
name: 0-shot |
|
- type: accuracy |
|
value: 66.54 |
|
name: 1-shot |
|
- type: accuracy |
|
value: 67.88 |
|
name: 3-shot |
|
- type: accuracy |
|
value: 69.3 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: OpenLLM-Ro/ro_hellaswag |
|
type: OpenLLM-Ro/ro_hellaswag |
|
metrics: |
|
- type: accuracy |
|
value: 57.69 |
|
name: Average accuracy |
|
- type: accuracy |
|
value: 56.12 |
|
name: 0-shot |
|
- type: accuracy |
|
value: 57.37 |
|
name: 1-shot |
|
- type: accuracy |
|
value: 57.92 |
|
name: 3-shot |
|
- type: accuracy |
|
value: 58.18 |
|
name: 5-shot |
|
- type: accuracy |
|
value: 58.85 |
|
name: 10-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: OpenLLM-Ro/ro_gsm8k |
|
type: OpenLLM-Ro/ro_gsm8k |
|
metrics: |
|
- type: accuracy |
|
value: 30.23 |
|
name: Average accuracy |
|
- type: accuracy |
|
value: 29.42 |
|
name: 1-shot |
|
- type: accuracy |
|
value: 30.02 |
|
name: 3-shot |
|
- type: accuracy |
|
value: 31.24 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: OpenLLM-Ro/ro_truthfulqa |
|
type: OpenLLM-Ro/ro_truthfulqa |
|
metrics: |
|
- type: accuracy |
|
value: 51.34 |
|
name: Average accuracy |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: LaRoSeDa_binary |
|
type: LaRoSeDa_binary |
|
metrics: |
|
- type: macro-f1 |
|
value: 97.52 |
|
name: Average macro-f1 |
|
- type: macro-f1 |
|
value: 97.43 |
|
name: 0-shot |
|
- type: macro-f1 |
|
value: 96.6 |
|
name: 1-shot |
|
- type: macro-f1 |
|
value: 97.9 |
|
name: 3-shot |
|
- type: macro-f1 |
|
value: 98.13 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: LaRoSeDa_multiclass |
|
type: LaRoSeDa_multiclass |
|
metrics: |
|
- type: macro-f1 |
|
value: 67.41 |
|
name: Average macro-f1 |
|
- type: macro-f1 |
|
value: 63.77 |
|
name: 0-shot |
|
- type: macro-f1 |
|
value: 68.91 |
|
name: 1-shot |
|
- type: macro-f1 |
|
value: 66.36 |
|
name: 3-shot |
|
- type: macro-f1 |
|
value: 70.61 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: LaRoSeDa_binary_finetuned |
|
type: LaRoSeDa_binary_finetuned |
|
metrics: |
|
- type: macro-f1 |
|
value: 94.15 |
|
name: Average macro-f1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: LaRoSeDa_multiclass_finetuned |
|
type: LaRoSeDa_multiclass_finetuned |
|
metrics: |
|
- type: macro-f1 |
|
value: 87.13 |
|
name: Average macro-f1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: WMT_EN-RO |
|
type: WMT_EN-RO |
|
metrics: |
|
- type: bleu |
|
value: 24.01 |
|
name: Average bleu |
|
- type: bleu |
|
value: 6.92 |
|
name: 0-shot |
|
- type: bleu |
|
value: 29.33 |
|
name: 1-shot |
|
- type: bleu |
|
value: 29.79 |
|
name: 3-shot |
|
- type: bleu |
|
value: 30.02 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: WMT_RO-EN |
|
type: WMT_RO-EN |
|
metrics: |
|
- type: bleu |
|
value: 27.36 |
|
name: Average bleu |
|
- type: bleu |
|
value: 4.5 |
|
name: 0-shot |
|
- type: bleu |
|
value: 30.3 |
|
name: 1-shot |
|
- type: bleu |
|
value: 36.96 |
|
name: 3-shot |
|
- type: bleu |
|
value: 37.7 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: WMT_EN-RO_finetuned |
|
type: WMT_EN-RO_finetuned |
|
metrics: |
|
- type: bleu |
|
value: 26.53 |
|
name: Average bleu |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: WMT_RO-EN_finetuned |
|
type: WMT_RO-EN_finetuned |
|
metrics: |
|
- type: bleu |
|
value: 40.36 |
|
name: Average bleu |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: XQuAD |
|
type: XQuAD |
|
metrics: |
|
- type: exact_match |
|
value: 39.43 |
|
name: Average exact_match |
|
- type: f1 |
|
value: 59.5 |
|
name: Average f1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: XQuAD_finetuned |
|
type: XQuAD_finetuned |
|
metrics: |
|
- type: exact_match |
|
value: 44.45 |
|
name: Average exact_match |
|
- type: f1 |
|
value: 59.76 |
|
name: Average f1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: STS |
|
type: STS |
|
metrics: |
|
- type: spearman |
|
value: 77.2 |
|
name: Average spearman |
|
- type: pearson |
|
value: 77.87 |
|
name: Average pearson |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: STS_finetuned |
|
type: STS_finetuned |
|
metrics: |
|
- type: spearman |
|
value: 85.8 |
|
name: Average spearman |
|
- type: pearson |
|
value: 86.05 |
|
name: Average pearson |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: XQuAD_EM |
|
type: XQuAD_EM |
|
metrics: |
|
- type: exact_match |
|
value: 4.45 |
|
name: 0-shot |
|
- type: exact_match |
|
value: 48.24 |
|
name: 1-shot |
|
- type: exact_match |
|
value: 52.03 |
|
name: 3-shot |
|
- type: exact_match |
|
value: 53.03 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: XQuAD_F1 |
|
type: XQuAD_F1 |
|
metrics: |
|
- type: f1 |
|
value: 26.08 |
|
name: 0-shot |
|
- type: f1 |
|
value: 68.4 |
|
name: 1-shot |
|
- type: f1 |
|
value: 71.92 |
|
name: 3-shot |
|
- type: f1 |
|
value: 71.6 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: STS_Spearman |
|
type: STS_Spearman |
|
metrics: |
|
- type: spearman |
|
value: 77.76 |
|
name: 1-shot |
|
- type: spearman |
|
value: 76.72 |
|
name: 3-shot |
|
- type: spearman |
|
value: 77.12 |
|
name: 5-shot |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: STS_Pearson |
|
type: STS_Pearson |
|
metrics: |
|
- type: pearson |
|
value: 77.83 |
|
name: 1-shot |
|
- type: pearson |
|
value: 77.64 |
|
name: 3-shot |
|
- type: pearson |
|
value: 78.13 |
|
name: 5-shot |
|
--- |
|
|
|
# vladciocan88/RoLlama3-8b-Instruct-2024-06-28-Q8_0-GGUF |
|
This model was converted to GGUF format from [`OpenLLM-Ro/RoLlama3-8b-Instruct-2024-06-28`](https://huggingface.co/OpenLLM-Ro/RoLlama3-8b-Instruct-2024-06-28) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space. |
|
Refer to the [original model card](https://huggingface.co/OpenLLM-Ro/RoLlama3-8b-Instruct-2024-06-28) for more details on the model. |
|
|
|
## Use with llama.cpp |
|
Install llama.cpp through brew (works on Mac and Linux) |
|
|
|
```bash |
|
brew install llama.cpp |
|
|
|
``` |
|
Invoke the llama.cpp server or the CLI. |
|
|
|
### CLI: |
|
```bash |
|
llama-cli --hf-repo vladciocan88/RoLlama3-8b-Instruct-2024-06-28-Q8_0-GGUF --hf-file rollama3-8b-instruct-2024-06-28-q8_0.gguf -p "The meaning to life and the universe is" |
|
``` |
|
|
|
### Server: |
|
```bash |
|
llama-server --hf-repo vladciocan88/RoLlama3-8b-Instruct-2024-06-28-Q8_0-GGUF --hf-file rollama3-8b-instruct-2024-06-28-q8_0.gguf -c 2048 |
|
``` |
|
|
|
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) 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 vladciocan88/RoLlama3-8b-Instruct-2024-06-28-Q8_0-GGUF --hf-file rollama3-8b-instruct-2024-06-28-q8_0.gguf -p "The meaning to life and the universe is" |
|
``` |
|
or |
|
``` |
|
./llama-server --hf-repo vladciocan88/RoLlama3-8b-Instruct-2024-06-28-Q8_0-GGUF --hf-file rollama3-8b-instruct-2024-06-28-q8_0.gguf -c 2048 |
|
``` |
|
|