Push model using huggingface_hub.
Browse files- README.md +9 -0
- config.json +28 -0
- model.safetensors +3 -0
README.md
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- pytorch_model_hub_mixin
|
4 |
+
- model_hub_mixin
|
5 |
+
---
|
6 |
+
|
7 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
|
8 |
+
- Library: [More Information Needed]
|
9 |
+
- Docs: [More Information Needed]
|
config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"attn_eq": false,
|
3 |
+
"attn_type": "trittention",
|
4 |
+
"autopad": true,
|
5 |
+
"causal_attn": true,
|
6 |
+
"d_head": 64,
|
7 |
+
"d_mlp": 2048,
|
8 |
+
"d_model": 768,
|
9 |
+
"d_vocab": 50257,
|
10 |
+
"debug": true,
|
11 |
+
"dropout": 0.1,
|
12 |
+
"dt_head": 64,
|
13 |
+
"freqs_cis": null,
|
14 |
+
"has_mlp": true,
|
15 |
+
"init_range": 0.02,
|
16 |
+
"is_gated": false,
|
17 |
+
"layer_norm_eps": 1e-05,
|
18 |
+
"look_backward": 1,
|
19 |
+
"mlp_type": "all",
|
20 |
+
"n_ctx": 128,
|
21 |
+
"n_heads": 12,
|
22 |
+
"n_layers": 1,
|
23 |
+
"nt_heads": 2,
|
24 |
+
"order_attn": true,
|
25 |
+
"pad_value": 0,
|
26 |
+
"window_size": 16,
|
27 |
+
"with_ln": true
|
28 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e37854604f6e82a0576c2c616a65ca55c4abdb23fcd57cfe7ec185aea0999440
|
3 |
+
size 336159192
|