Update README.md
Browse files
README.md
CHANGED
@@ -6,12 +6,12 @@ tags: []
|
|
6 |
from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer
|
7 |
import torch
|
8 |
|
9 |
-
tokenizer = AutoTokenizer.from_pretrained("
|
10 |
tokenizer.pad_token = tokenizer.eos_token
|
11 |
|
12 |
-
model = MambaForCausalLM.from_pretrained("
|
13 |
model.config.use_cache = True
|
14 |
-
input_ids = tokenizer("Hey how are you doing?", return_tensors= "pt")["input_ids"]
|
15 |
|
16 |
out = model.generate(input_ids, max_new_tokens=10)
|
17 |
print(tokenizer.batch_decode(out))
|
|
|
6 |
from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer
|
7 |
import torch
|
8 |
|
9 |
+
tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neox-20b", padding_side = "left")
|
10 |
tokenizer.pad_token = tokenizer.eos_token
|
11 |
|
12 |
+
model = MambaForCausalLM.from_pretrained("state-spaces/mamba-130m", vocab_size=50280, num_hidden_layers=24, torch_dtype=torch.float32)
|
13 |
model.config.use_cache = True
|
14 |
+
input_ids = tokenizer(["Hey how are you doing?", "Explain how soy sauce is made"], padding=True, return_tensors= "pt")["input_ids"]
|
15 |
|
16 |
out = model.generate(input_ids, max_new_tokens=10)
|
17 |
print(tokenizer.batch_decode(out))
|