Fix Llama language typo in bonus materials (#597)

Author: rasbt

ch05/07_gpt_to_llama/standalone-llama32-mem-opt.ipynb

@@ -53,7 +53,7 @@
     "  \n",
     "- About the code:\n",
     "  - all code is my own code, mapping the Llama 3 architecture onto the model code implemented in my [Build A Large Language Model (From Scratch)](http://mng.bz/orYv) book; the code is released under a permissive open-source Apache 2.0 license (see [LICENSE.txt](https://github.com/rasbt/LLMs-from-scratch/blob/main/LICENSE.txt))\n",
-    "  - the tokenizer code is inspired by the original [Llama 3 tokenizer code](https://github.com/meta-llama/llama3/blob/main/llama/tokenizer.py), which Meta AI used to to extends the Tiktoken GPT-4 tokenizer\n",
+    "  - the tokenizer code is inspired by the original [Llama 3 tokenizer code](https://github.com/meta-llama/llama3/blob/main/llama/tokenizer.py), which Meta AI used to extend the Tiktoken GPT-4 tokenizer\n",
     "  - the RoPE rescaling section is inspired by the [_compute_llama3_parameters function](https://github.com/huggingface/transformers/blob/5c1027bf09717f664b579e01cbb8ec3ef5aeb140/src/transformers/modeling_rope_utils.py#L329-L347) in the `transformers` library"
    ]
   },

ch05/07_gpt_to_llama/standalone-llama32.ipynb

@@ -50,7 +50,7 @@
     "  \n",
     "- About the code:\n",
     "  - all code is my own code, mapping the Llama 3 architecture onto the model code implemented in my [Build A Large Language Model (From Scratch)](http://mng.bz/orYv) book; the code is released under a permissive open-source Apache 2.0 license (see [LICENSE.txt](https://github.com/rasbt/LLMs-from-scratch/blob/main/LICENSE.txt))\n",
-    "  - the tokenizer code is inspired by the original [Llama 3 tokenizer code](https://github.com/meta-llama/llama3/blob/main/llama/tokenizer.py), which Meta AI used to to extends the Tiktoken GPT-4 tokenizer\n",
+    "  - the tokenizer code is inspired by the original [Llama 3 tokenizer code](https://github.com/meta-llama/llama3/blob/main/llama/tokenizer.py), which Meta AI used to extend the Tiktoken GPT-4 tokenizer\n",
     "  - the RoPE rescaling section is inspired by the [_compute_llama3_parameters function](https://github.com/huggingface/transformers/blob/5c1027bf09717f664b579e01cbb8ec3ef5aeb140/src/transformers/modeling_rope_utils.py#L329-L347) in the `transformers` library"
    ]
   },