ganqu commited on
Commit
0557982
1 Parent(s): d962e51

fix length & rope

Browse files
Files changed (1) hide show
  1. config.json +2 -2
config.json CHANGED
@@ -11,7 +11,7 @@
11
  "hidden_size": 8192,
12
  "initializer_range": 0.02,
13
  "intermediate_size": 28672,
14
- "max_position_embeddings": 4096,
15
  "model_type": "llama",
16
  "num_attention_heads": 64,
17
  "num_hidden_layers": 80,
@@ -19,7 +19,7 @@
19
  "pretraining_tp": 1,
20
  "rms_norm_eps": 1e-05,
21
  "rope_scaling": null,
22
- "rope_theta": 10000,
23
  "tie_word_embeddings": false,
24
  "torch_dtype": "float16",
25
  "transformers_version": "4.35.0",
 
11
  "hidden_size": 8192,
12
  "initializer_range": 0.02,
13
  "intermediate_size": 28672,
14
+ "max_position_embeddings": 16384,
15
  "model_type": "llama",
16
  "num_attention_heads": 64,
17
  "num_hidden_layers": 80,
 
19
  "pretraining_tp": 1,
20
  "rms_norm_eps": 1e-05,
21
  "rope_scaling": null,
22
+ "rope_theta": 1000000,
23
  "tie_word_embeddings": false,
24
  "torch_dtype": "float16",
25
  "transformers_version": "4.35.0",