Update app.py

app.py

@@ -18,7 +18,7 @@ from transformers import (
 
 
 model_name = "mosaicml/mpt-7b-chat"
-max_new_tokens = 1536
+max_new_tokens = 7000
 
 # # small testing model:
 # model_name = "gpt2"
@@ -33,7 +33,7 @@ m = AutoModelForCausalLM.from_pretrained(
     torch_dtype=torch.bfloat16,
     trust_remote_code=True,
     use_auth_token=auth_token,
-    max_seq_len=8192,
+    max_seq_len=10000,
 ).cuda()
 tok = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True, use_auth_token=auth_token)