We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent 24b2980 commit 0a1454eCopy full SHA for 0a1454e
configs/gsa_1B.json
@@ -0,0 +1,29 @@
1
+{
2
+ "bos_token_id": 1,
3
+ "conv_size": 4,
4
+ "eos_token_id": 2,
5
+ "expand_k": 1,
6
+ "expand_v": 1,
7
+ "elementwise_affine": false,
8
+ "feature_map": "swish",
9
+ "fuse_cross_entropy": true,
10
+ "fuse_norm": true,
11
+ "gate_logit_normalizer": 8,
12
+ "hidden_act": "swish",
13
+ "hidden_ratio": 4,
14
+ "hidden_size": 2048,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": null,
17
+ "model_type": "gsa",
18
+ "num_heads": 4,
19
+ "num_hidden_layers": 24,
20
+ "num_slots": 64,
21
+ "norm_eps": 1e-06,
22
+ "share_conv_kernel": true,
23
+ "tie_word_embeddings": false,
24
+ "use_cache": true,
25
+ "use_norm": true,
26
+ "use_output_gate": true,
27
+ "use_rope": false,
28
+ "use_short_conv": false
29
+}
0 commit comments