Initial kimetsu-style tiny GPT
Browse files- config.json +2 -2
- model.safetensors +2 -2
- training_args.bin +1 -1
config.json
CHANGED
|
@@ -11,12 +11,12 @@
|
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"layer_norm_epsilon": 1e-05,
|
| 13 |
"model_type": "gpt2",
|
| 14 |
-
"n_ctx":
|
| 15 |
"n_embd": 256,
|
| 16 |
"n_head": 4,
|
| 17 |
"n_inner": null,
|
| 18 |
"n_layer": 2,
|
| 19 |
-
"n_positions":
|
| 20 |
"reorder_and_upcast_attn": false,
|
| 21 |
"resid_pdrop": 0.1,
|
| 22 |
"scale_attn_by_inverse_layer_idx": false,
|
|
|
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"layer_norm_epsilon": 1e-05,
|
| 13 |
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 256,
|
| 15 |
"n_embd": 256,
|
| 16 |
"n_head": 4,
|
| 17 |
"n_inner": null,
|
| 18 |
"n_layer": 2,
|
| 19 |
+
"n_positions": 256,
|
| 20 |
"reorder_and_upcast_attn": false,
|
| 21 |
"resid_pdrop": 0.1,
|
| 22 |
"scale_attn_by_inverse_layer_idx": false,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc0317349411b7f52ba711c4d5594f3d5177aa8b185e01346dab9a33fe4d0a57
|
| 3 |
+
size 39353008
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5777
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ea5e711c9c44476ebdae23ca1f2093187c4c2dcfafbe692f8237bb3fd506bf9
|
| 3 |
size 5777
|