Upload FalconForDistill
Browse files- config.json +3 -3
- generation_config.json +1 -1
- pytorch_model.bin +2 -2
config.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"alibi": true,
|
4 |
"apply_residual_connection_post_layernorm": false,
|
5 |
"architectures": [
|
6 |
-
"
|
7 |
],
|
8 |
"attention_dropout": 0.0,
|
9 |
"auto_map": {
|
@@ -16,7 +16,7 @@
|
|
16 |
},
|
17 |
"bias": true,
|
18 |
"bos_token_id": 1,
|
19 |
-
"eos_token_id":
|
20 |
"hidden_dropout": 0.0,
|
21 |
"hidden_size": 2048,
|
22 |
"initializer_range": 0.02,
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "model/falcon/falcon-rw-1b-bf16-train_batch=4-train=rw_code_falcon_dataset",
|
3 |
"alibi": true,
|
4 |
"apply_residual_connection_post_layernorm": false,
|
5 |
"architectures": [
|
6 |
+
"FalconForDistill"
|
7 |
],
|
8 |
"attention_dropout": 0.0,
|
9 |
"auto_map": {
|
|
|
16 |
},
|
17 |
"bias": true,
|
18 |
"bos_token_id": 1,
|
19 |
+
"eos_token_id": null,
|
20 |
"hidden_dropout": 0.0,
|
21 |
"hidden_size": 2048,
|
22 |
"initializer_range": 0.02,
|
generation_config.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
"_from_model_config": true,
|
3 |
"bos_token_id": 1,
|
4 |
-
"
|
5 |
"transformers_version": "4.34.1"
|
6 |
}
|
|
|
1 |
{
|
2 |
"_from_model_config": true,
|
3 |
"bos_token_id": 1,
|
4 |
+
"pad_token_id": 50256,
|
5 |
"transformers_version": "4.34.1"
|
6 |
}
|
pytorch_model.bin
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:d432331081341fc4edf8f0bb76f72301441d0723a914cda9dd5f3da6f90367d9
|
3 |
+
size 5246704278
|