Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- coconut
|
| 4 |
+
- gpt2
|
| 5 |
+
- reasoning
|
| 6 |
+
base_model: openai-community/gpt2
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# gpt2-coconut-checkpoint13
|
| 10 |
+
|
| 11 |
+
CoCoNut (Chain of Continuous Thought) trained checkpoint 13 based on GPT-2.
|
| 12 |
+
|
| 13 |
+
## Training
|
| 14 |
+
- Base model: `openai-community/gpt2`
|
| 15 |
+
- Training method: CoCoNut (continuous latent reasoning)
|
| 16 |
+
- Dataset: GSM8K
|
| 17 |
+
|
| 18 |
+
## Usage
|
| 19 |
+
```python
|
| 20 |
+
import torch
|
| 21 |
+
checkpoint = torch.load("pytorch_model.bin")
|
| 22 |
+
```
|