Update README.md
Browse files
README.md
CHANGED
@@ -3,6 +3,9 @@ tags:
|
|
3 |
- llm-rs
|
4 |
- ggml
|
5 |
pipeline_tag: text-generation
|
|
|
|
|
|
|
6 |
---
|
7 |
# GGML converted versions of [Together](https://huggingface.co/togethercomputer)'s RedPajama models
|
8 |
|
@@ -27,3 +30,34 @@ The training was done on 3,072 V100 GPUs provided as part of the INCITE 2023 pro
|
|
27 |
| [RedPajama-INCITE-Instruct-3B-v1-f16.bin](https://huggingface.co/Rustformers/redpajama-ggml/blob/main/RedPajama-INCITE-Instruct-3B-v1-f16.bin) | [RedPajama-INCITE-Instruct](https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1) | fp16 | GGML |
|
28 |
| [RedPajama-INCITE-Instruct-3B-v1-q4_0-ggjt.bin](https://huggingface.co/Rustformers/redpajama-ggml/blob/main/RedPajama-INCITE-Instruct-3B-v1-q4_0-ggjt.bin) | [RedPajama-INCITE-Instruct](https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1) | int4 | GGJT |
|
29 |
| [RedPajama-INCITE-Instruct-3B-v1-q4_0.bin](https://huggingface.co/Rustformers/redpajama-ggml/blob/main/RedPajama-INCITE-Instruct-3B-v1-q4_0.bin) | [RedPajama-INCITE-Instruct](https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1) | int4 | GGML |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
- llm-rs
|
4 |
- ggml
|
5 |
pipeline_tag: text-generation
|
6 |
+
license: apache-2.0
|
7 |
+
language:
|
8 |
+
- en
|
9 |
---
|
10 |
# GGML converted versions of [Together](https://huggingface.co/togethercomputer)'s RedPajama models
|
11 |
|
|
|
30 |
| [RedPajama-INCITE-Instruct-3B-v1-f16.bin](https://huggingface.co/Rustformers/redpajama-ggml/blob/main/RedPajama-INCITE-Instruct-3B-v1-f16.bin) | [RedPajama-INCITE-Instruct](https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1) | fp16 | GGML |
|
31 |
| [RedPajama-INCITE-Instruct-3B-v1-q4_0-ggjt.bin](https://huggingface.co/Rustformers/redpajama-ggml/blob/main/RedPajama-INCITE-Instruct-3B-v1-q4_0-ggjt.bin) | [RedPajama-INCITE-Instruct](https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1) | int4 | GGJT |
|
32 |
| [RedPajama-INCITE-Instruct-3B-v1-q4_0.bin](https://huggingface.co/Rustformers/redpajama-ggml/blob/main/RedPajama-INCITE-Instruct-3B-v1-q4_0.bin) | [RedPajama-INCITE-Instruct](https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1) | int4 | GGML |
|
33 |
+
|
34 |
+
## Usage
|
35 |
+
|
36 |
+
### Python via [llm-rs](https://github.com/LLukas22/llm-rs-python):
|
37 |
+
|
38 |
+
#### Installation
|
39 |
+
Via pip: `pip install llm-rs`
|
40 |
+
|
41 |
+
#### Run inference
|
42 |
+
```python
|
43 |
+
from llm_rs import AutoModel
|
44 |
+
|
45 |
+
#Load the model, define any model you like from the list above as the `model_file`
|
46 |
+
model = AutoModel.from_pretrained("Rustformers/redpajama-ggml",model_file="RedPajama-INCITE-Base-3B-v1-q4_0-ggjt.bin")
|
47 |
+
|
48 |
+
#Generate
|
49 |
+
print(model.generate("The meaning of life is"))
|
50 |
+
```
|
51 |
+
|
52 |
+
### Rust via [Rustformers/llm](https://github.com/rustformers/llm):
|
53 |
+
|
54 |
+
#### Installation
|
55 |
+
```
|
56 |
+
git clone --recurse-submodules [email protected]:rustformers/llm.git
|
57 |
+
cargo build --release
|
58 |
+
```
|
59 |
+
|
60 |
+
#### Run inference
|
61 |
+
```
|
62 |
+
cargo run --release -- gptneox infer -m path/to/model.bin -p "Tell me how cool the Rust programming language is:"
|
63 |
+
```
|