Upload model - 35000 iterations, loss: 3.4640
Browse files- config.json +1 -1
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -9,7 +9,7 @@
|
|
| 9 |
"n_layer": 8,
|
| 10 |
"n_head": 8,
|
| 11 |
"n_inner": 1536,
|
| 12 |
-
"activation_function": "
|
| 13 |
"resid_pdrop": 0.1,
|
| 14 |
"embd_pdrop": 0.1,
|
| 15 |
"attn_pdrop": 0.1,
|
|
|
|
| 9 |
"n_layer": 8,
|
| 10 |
"n_head": 8,
|
| 11 |
"n_inner": 1536,
|
| 12 |
+
"activation_function": "gelu",
|
| 13 |
"resid_pdrop": 0.1,
|
| 14 |
"embd_pdrop": 0.1,
|
| 15 |
"attn_pdrop": 0.1,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 211972024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5af8ee2b8124c95ea8667396f3d525176f9e40b8d85f6331034a8bba7245c3e3
|
| 3 |
size 211972024
|