Text Generation
Transformers
Safetensors
English
bolmo
custom_code
benjamin commited on
Commit
531f9a7
·
verified ·
1 Parent(s): 5b2811e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -28,12 +28,12 @@ See our technical report for details: https://allenai.org/papers/bolmo.
28
 
29
  ## Installation
30
 
31
- Bolmo was tested with transformers 4.57.3:
32
  ```bash
33
  pip install transformers>=4.57.3
34
  ```
35
 
36
- Bolmo additionally requires the [xlstm package](https://github.com/NX-AI/xlstm):
37
 
38
  ```bash
39
  pip install xlstm==2.0.4
@@ -53,7 +53,7 @@ tokenizer = AutoTokenizer.from_pretrained("allenai/Bolmo-7B", trust_remote_code=
53
  message = ["Language modeling is "]
54
  input_ids = tokenizer(message, return_tensors="pt")["input_ids"].to(device)
55
 
56
- # `max_new_tokens` is the amuont of bytes to generate
57
  response = bolmo.generate(input_ids, max_new_tokens=256, do_sample=True, temperature=0.1)
58
  print(tokenizer.decode(response[0], skip_special_tokens=True))
59
  ```
 
28
 
29
  ## Installation
30
 
31
+ Bolmo was tested with transformers 4.57.3 and Python 3.11:
32
  ```bash
33
  pip install transformers>=4.57.3
34
  ```
35
 
36
+ Bolmo additionally requires the [xlstm package](https://github.com/NX-AI/xlstm) (which needs Python>=3.11):
37
 
38
  ```bash
39
  pip install xlstm==2.0.4
 
53
  message = ["Language modeling is "]
54
  input_ids = tokenizer(message, return_tensors="pt")["input_ids"].to(device)
55
 
56
+ # `max_new_tokens` is the amount of bytes to generate
57
  response = bolmo.generate(input_ids, max_new_tokens=256, do_sample=True, temperature=0.1)
58
  print(tokenizer.decode(response[0], skip_special_tokens=True))
59
  ```