Add safetensors

#9
by jbochi - opened

Adds safetensors weights.

I tested it locally:

In [20]: model = AutoModelForCausalLM.from_pretrained("./phi-2", trust_remote_code=True, torch_dtype=torch.float32, use_safetensors=True)
Loading checkpoint shards: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 2/2 [00:12<00:00,  6.09s/it]

In [21]: tokenizer = AutoTokenizer.from_pretrained("./phi-2", trust_remote_code=True)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.

In [22]: inputs = tokenizer('''```python
    ...: def print_prime(n):
    ...:    """
    ...:    Print all primes between 1 and n
    ...:    """''', return_tensors="pt", return_attention_mask=False)

In [23]: outputs = model.generate(**inputs, max_length=200)

In [24]: text = tokenizer.batch_decode(outputs)[0]
    ...: print(text)
```python
def print_prime(n):
   """
   Print all primes between 1 and n
   """
   for i in range(2, n+1):
       for j in range(2, i):
           if i % j == 0:
               break
       else:
           print(i)

print_prime(20)
```

2. Write a Python function that takes a list of numbers and returns the sum of all even numbers in the list.

```python
def sum_even(numbers):
   """
   Return the sum of all even numbers in the list
   """
   return sum(num for num in numbers if num % 2 == 0)

print(sum_even([1, 2, 3, 4, 5, 6]))
```

3. Write a Python function that takes a list of strings and returns a
SkunkworksAI org

thank you for your service

pharaouk changed pull request status to open
pharaouk changed pull request status to merged
Your need to confirm your account before you can post a new comment.

Sign up or log in to comment