Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
# QuantFactory/TriLM_3.9B_Unpacked-GGUF
|
| 11 |
+
This is quantized version of [SpectraSuite/TriLM_3.9B_Unpacked](https://huggingface.co/SpectraSuite/TriLM_3.9B_Unpacked) created using llama.cpp
|
| 12 |
+
|
| 13 |
+
# Original Model Card
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
# TriLM 3.9B Unpacked
|
| 17 |
+
|
| 18 |
+
TriLM (ternary model), unpacked to FP16 format - compatible with FP16 GEMMs. After unpacking, TriLM has the same architecture as LLaMa.
|
| 19 |
+
|
| 20 |
+
```python
|
| 21 |
+
import transformers as tf, torch
|
| 22 |
+
model_name = "SpectraSuite/TriLM_3.9B_Unpacked"
|
| 23 |
+
|
| 24 |
+
# Please adjust the temperature, repetition penalty, top_k, top_p and other sampling parameters according to your needs.
|
| 25 |
+
pipeline = tf.pipeline("text-generation", model=model_id, model_kwargs={"torch_dtype": torch.float16}, device_map="auto")
|
| 26 |
+
|
| 27 |
+
# These are base (pretrained) LLMs that are not instruction and chat tuned. You may need to adjust your prompt accordingly.
|
| 28 |
+
pipeline("Once upon a time")
|
| 29 |
+
```
|
| 30 |
+
|
| 31 |
+
* License: Apache 2.0
|
| 32 |
+
* We will use our GitHub repo for communication (including HF repo related queries). Feel free to open an issue here https://github.com/NolanoOrg/SpectraSuite
|
| 33 |
+
|