Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
tags:
|
| 6 |
+
- pytorch
|
| 7 |
+
- causal-lm
|
| 8 |
+
- custom
|
| 9 |
+
license: apache-2.0
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Sky610TX
|
| 13 |
+
|
| 14 |
+
**Sky610TX** is a custom 610 Million parameter language model trained on the Sky dataset.
|
| 15 |
+
|
| 16 |
+
## Model Details
|
| 17 |
+
- **Architecture:** GPT-2 Style (Custom Ascendant Config)
|
| 18 |
+
- **Parameters:** ~610 Million
|
| 19 |
+
- **Context Window:** 1024 Tokens
|
| 20 |
+
- **Was made as a test, new 1B parameter model coming soon!**
|
| 21 |
+
|
| 22 |
+
## How to Use
|
| 23 |
+
```python
|
| 24 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 25 |
+
|
| 26 |
+
model = AutoModelForCausalLM.from_pretrained("8BitStudio/Sky610TX")
|
| 27 |
+
tokenizer = AutoTokenizer.from_pretrained("8BitStudio/Sky610TX")
|
| 28 |
+
|
| 29 |
+
input_text = "User: Hello\nAssistant:"
|
| 30 |
+
inputs = tokenizer(input_text, return_tensors="pt")
|
| 31 |
+
outputs = model.generate(**inputs, max_new_tokens=50)
|
| 32 |
+
print(tokenizer.decode(outputs[0]))
|