NanoGPT-SIGReg strong - Val Loss: 7.9404
Browse files- README.md +11 -0
- config.json +10 -0
- model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: nanogpt-sigreg
|
| 3 |
+
tags:
|
| 4 |
+
- model_hub_mixin
|
| 5 |
+
- pytorch_model_hub_mixin
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
|
| 9 |
+
- Code: [More Information Needed]
|
| 10 |
+
- Paper: [More Information Needed]
|
| 11 |
+
- Docs: [More Information Needed]
|
config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"depth": 24,
|
| 3 |
+
"dim": 640,
|
| 4 |
+
"heads": 10,
|
| 5 |
+
"kv_heads": 2,
|
| 6 |
+
"reg_mode": "strong",
|
| 7 |
+
"sigreg_alpha": 0.01,
|
| 8 |
+
"sketch_dim": 64,
|
| 9 |
+
"vocab_size": 50304
|
| 10 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f71a0429e742ae34fc9d1bd40515021afa7d1fb2da18d9a2d19cde5a8e1e4655
|
| 3 |
+
size 460287288
|