Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: sapienzanlp/Minerva-3B-base-v1.0
|
| 3 |
+
datasets:
|
| 4 |
+
- uonlp/CulturaX
|
| 5 |
+
language:
|
| 6 |
+
- it
|
| 7 |
+
- en
|
| 8 |
+
license: apache-2.0
|
| 9 |
+
pipeline_tag: text-generation
|
| 10 |
+
tags:
|
| 11 |
+
- pretrained
|
| 12 |
+
- llama-cpp
|
| 13 |
+
- gguf-my-repo
|
| 14 |
+
inference:
|
| 15 |
+
parameters:
|
| 16 |
+
temperature: 0.5
|
| 17 |
+
do_sample: true
|
| 18 |
+
widget:
|
| 19 |
+
- text: 'La capitale dell''Italia è '
|
| 20 |
+
example_title: Example 1
|
| 21 |
+
- text: 'Nel mezzo del cammin di nostra vita '
|
| 22 |
+
example_title: Example 2
|
| 23 |
+
- text: 'Una cena senza vino è come '
|
| 24 |
+
example_title: Example 3
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
# antoste/Minerva-3B-base-v1.0-IQ3_M-GGUF
|
| 28 |
+
This model was converted to GGUF format from [`sapienzanlp/Minerva-3B-base-v1.0`](https://huggingface.co/sapienzanlp/Minerva-3B-base-v1.0) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
| 29 |
+
Refer to the [original model card](https://huggingface.co/sapienzanlp/Minerva-3B-base-v1.0) for more details on the model.
|
| 30 |
+
|
| 31 |
+
## Use with llama.cpp
|
| 32 |
+
Install llama.cpp through brew (works on Mac and Linux)
|
| 33 |
+
|
| 34 |
+
```bash
|
| 35 |
+
brew install llama.cpp
|
| 36 |
+
|
| 37 |
+
```
|
| 38 |
+
Invoke the llama.cpp server or the CLI.
|
| 39 |
+
|
| 40 |
+
### CLI:
|
| 41 |
+
```bash
|
| 42 |
+
llama-cli --hf-repo antoste/Minerva-3B-base-v1.0-IQ3_M-GGUF --hf-file minerva-3b-base-v1.0-iq3_m-imat.gguf -p "The meaning to life and the universe is"
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
### Server:
|
| 46 |
+
```bash
|
| 47 |
+
llama-server --hf-repo antoste/Minerva-3B-base-v1.0-IQ3_M-GGUF --hf-file minerva-3b-base-v1.0-iq3_m-imat.gguf -c 2048
|
| 48 |
+
```
|
| 49 |
+
|
| 50 |
+
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.
|
| 51 |
+
|
| 52 |
+
Step 1: Clone llama.cpp from GitHub.
|
| 53 |
+
```
|
| 54 |
+
git clone https://github.com/ggerganov/llama.cpp
|
| 55 |
+
```
|
| 56 |
+
|
| 57 |
+
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).
|
| 58 |
+
```
|
| 59 |
+
cd llama.cpp && LLAMA_CURL=1 make
|
| 60 |
+
```
|
| 61 |
+
|
| 62 |
+
Step 3: Run inference through the main binary.
|
| 63 |
+
```
|
| 64 |
+
./llama-cli --hf-repo antoste/Minerva-3B-base-v1.0-IQ3_M-GGUF --hf-file minerva-3b-base-v1.0-iq3_m-imat.gguf -p "The meaning to life and the universe is"
|
| 65 |
+
```
|
| 66 |
+
or
|
| 67 |
+
```
|
| 68 |
+
./llama-server --hf-repo antoste/Minerva-3B-base-v1.0-IQ3_M-GGUF --hf-file minerva-3b-base-v1.0-iq3_m-imat.gguf -c 2048
|
| 69 |
+
```
|