[DOCS] Example for LogitsProcessor class

src/transformers/generation/logits_process.py

@@ -193,12 +193,49 @@ def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor) -> to
 
 class RepetitionPenaltyLogitsProcessor(LogitsProcessor):
     r"""
-    [`LogitsProcessor`] enforcing an exponential penalty on repeated sequences.
+    [`LogitsProcessor`] prevents the repetition of previous tokens through an exponential penalty. 
+    This technique invloves adjusting the scores assigned to previously generated tokens, discouraging repetition. 
+    This technique shares some similarities with coverage mechanisms and other aimed at reducing repetition. 
+    During the text generation process, the probability distribution for the next token is determined using a formula that incorporates 
+    token scores based on their occurrence in the generated sequence. Tokens with higher scores are less likely to be selected. 
+    The formula takes into account a fair bunch of variables which can be further seen in the [paper](https://arxiv.org/pdf/1909.05858.pdf).
+
+    <Tip>
+
+    It is worth noting that the effectiveness of penalized sampling relies on the model's ability to learn a reliable distribution of 
+    tokens during its training.
+
+    </Tip>
 
     Args:
         repetition_penalty (`float`):
-            The parameter for repetition penalty. 1.0 means no penalty. See [this
-            paper](https://arxiv.org/pdf/1909.05858.pdf) for more details.
+            The parameter for repetition penalty. 1.0 means no penalty. According to the paper a penalty of around 1.2 
+            yields a good balance between truthful generation and lack of repetition. 
+            See [this paper](https://arxiv.org/pdf/1909.05858.pdf) for more details.
+
+    Examples:
+
+    ```python
+    >>> from transformers import AutoTokenizer, AutoModelForCausalLM
+
+    >>> # Initializing the model and tokenizer for it 
+    >>> model = AutoModelForCausalLM.from_pretrained("gpt2")
+    >>> # add_prefix_space - Adds an initial space to the input. This allows to treat the leading word just as any other word 
+    >>> #(Not important for this demonstration)
+    >>> tokenizer = AutoTokenizer.from_pretrained("gpt2", add_prefix_space = True)
+    >>> inputs = tokenizer(["I'm not going to"], return_tensors="pt")
+
+    >>> # This shows a normal generate without any specific parameters 
+    >>> summary_ids = model.generate(inputs["input_ids"], max_length=20)
+    >>> print(tokenizer.batch_decode(summary_ids, skip_special_tokens=True)[0])
+    I'm not going to lie, I'm not going to lie. I'm not going to lie
+
+    >>> # This generates a penalty for repeated tokens
+    >>> penalized_ids = model.generate(inputs["input_ids"], max_length=20, repetition_penalty=1.2)
+    >>> print(tokenizer.batch_decode(biased_ids, skip_special_tokens=True)[0])
+    I'm not going to lie, I was really excited about this. It's a great game
+
+    ```
     """
 
     def __init__(self, penalty: float):