Now, let's run a baby Llama 2 model in OCaml. You need a model checkpoint. Download this 15M parameter model trained on the TinyStories dataset (~60MB download):
Once upon a time, there was a little girl named Lily. She loved to play outside in the sunshine. One day, she saw a big, red ball in the sky. It was the sun! She thought it was so pretty. Lily wanted to play with the ball, but it was too high up in the sky. She tried to jump and reach it, but she couldn't. Then, she had an idea. She would use a stick to knock the ball down.
Once upon a time, there was a little girl named Lily. She loved to play outside in the sunshine. One day, she saw a big, yellow flower in the garden. It was a sunflower! Lily thought it was the most beautiful flower she had ever seen. Lily's mom came outside and saw the sunflower too. "Wow, that's a big flower!" she said. "Let's pick it and put it in a vase." Lily was so excited to have the sunflower in her room.
In the wild jungle, there was a zebra. He loved to hide in the bushes. One day, he was running around looking for a good place to hide when he noticed a small hut. He ran over to it and peeked inside. Inside he could see a family of foxes playing around. The zebra stayed still and watched them. The foxes noticed the zebra and ran away. The zebra was sad because he had no one to play with. He then noticed a tree full of leaves and ran over to it to hide behind its branches.