Commit
•
46c3dec
1
Parent(s):
32d98f1
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
tags:
|
| 4 |
+
- pytorch
|
| 5 |
+
- gpt2
|
| 6 |
+
- language-model
|
| 7 |
+
pipeline_tag: text-generation
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# GPT-X Model
|
| 11 |
+
|
| 12 |
+
This model was trained using the GPT-X framework.
|
| 13 |
+
|
| 14 |
+
## Model Architecture
|
| 15 |
+
|
| 16 |
+
- Layers: 12
|
| 17 |
+
- Attention Heads: 12
|
| 18 |
+
- Hidden Size: 768
|
| 19 |
+
- Vocabulary Size: 50257
|
| 20 |
+
- Maximum Sequence Length: 1024
|
| 21 |
+
- Model Type: base
|
| 22 |
+
|
| 23 |
+
## Training Details
|
| 24 |
+
|
| 25 |
+
- Batch Size: 524288
|
| 26 |
+
- Learning Rate: 0.0006
|
| 27 |
+
- Weight Decay: 0.0
|
| 28 |
+
- Mixed Precision: True
|
| 29 |
+
- Optimizer: muon
|