We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent bec61f0 commit 7af5591Copy full SHA for 7af5591
configs/nsa_340M.json
@@ -1,13 +1,5 @@
1
{
2
"attention_bias": false,
3
- "architectures": [
4
- "NSAForCausalLM"
5
- ],
6
- "auto_map": {
7
- "AutoConfig": "native_sparse_attention.NSAConfig",
8
- "AutoModel": "native_sparse_attention.NSAModel",
9
- "AutoModelForCausalLM": "native_sparse_attention.NSAForCausalLM"
10
- },
11
"bos_token_id": 1,
12
"eos_token_id": 2,
13
"fuse_cross_entropy": true,
@@ -27,4 +19,5 @@
27
19
"tie_word_embeddings": false,
28
20
"use_cache": true,
29
21
"vocab_size": 32000
30
-}
22
+}
23
+
0 commit comments