Patch checkpoint to transformers main format (convert_olmo_hybrid_hf_to_main)
Browse files- config.json +2 -3
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"architectures": [
|
| 3 |
-
"
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
| 6 |
"attention_dropout": 0.0,
|
|
@@ -49,10 +49,9 @@
|
|
| 49 |
"linear_key_head_dim": 96,
|
| 50 |
"linear_num_key_heads": 30,
|
| 51 |
"linear_num_value_heads": 30,
|
| 52 |
-
"linear_use_gate": true,
|
| 53 |
"linear_value_head_dim": 192,
|
| 54 |
"max_position_embeddings": 32768,
|
| 55 |
-
"model_type": "
|
| 56 |
"num_attention_heads": 30,
|
| 57 |
"num_hidden_layers": 32,
|
| 58 |
"num_key_value_heads": 30,
|
|
|
|
| 1 |
{
|
| 2 |
"architectures": [
|
| 3 |
+
"OlmoHybridForCausalLM"
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
| 6 |
"attention_dropout": 0.0,
|
|
|
|
| 49 |
"linear_key_head_dim": 96,
|
| 50 |
"linear_num_key_heads": 30,
|
| 51 |
"linear_num_value_heads": 30,
|
|
|
|
| 52 |
"linear_value_head_dim": 192,
|
| 53 |
"max_position_embeddings": 32768,
|
| 54 |
+
"model_type": "olmo_hybrid",
|
| 55 |
"num_attention_heads": 30,
|
| 56 |
"num_hidden_layers": 32,
|
| 57 |
"num_key_value_heads": 30,
|
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 sha256:40db118a86509576789d22a89dc09db3a91a0963f602c44c76d6ca678f58a324
|
| 3 |
+
size 14861802680
|