5 ms·
Each character made the cut, but the word "strawberry" is a single token, and that single token is what the model gets as input. When humans read some text, the
by kadushka 1y ago
Each character made the cut, but the word "strawberry" is a single token, and that single token is what the model gets as input. When humans read some text, they can see each individual character in the word "strawberry" everytime they see that word. LLMs don't see individual characters when they process input text containing the word "strawberry". They can only learn the spelling if some text explicitly maps "strawberry" to the sequence of characters s t r a w b e r r y. My guess is there are not enough of such mappings present in the training dataset for the model to learn it well.
- nl 1y ago> the word "strawberry" is a single token, and that single token is what the model gets as input. This is incorrect. strawberry is actually 4 tokens (at least for GPT but most LLM are similar). See https://platform.openai.com/tokenizer https://platform.openai.com/tokenizer
- kadushka 1y agoI got 3 tokens: st, raw, and berry. My point still stands: processing "berry" as a single token does not allow the model to learn its spelling directly, the way human readers do. It still has to rely on an explicit mapping of the word "berry" to b e r r y explained in some text in the training dataset. If that explanation is not present in the training data, it cannot learn the spelling - in principle.
- brookst 1y agoExactly. If “st” is 123, “raw” is 456, “berry” is 789, and “r” is 17… it makes little sense to ask the models to count the [17]’s in [123,466,789]: it demands an awareness of the abstraction that does not exist. To the extent the knowledge is there it’s from data in the input corpus, not direct examination of the text or tokens in the prompt.
- asadotzler 1y agoSo much for generalized intelligence, I guess.
- kadushka 1y agoIs a human who never learned how to read not generally intelligent?
- boroboro4 1y agoThe fact the word ends up being 1 token doesn’t mean model can’t track individual characters in it. The model transforms token into vector (of multiple thousands dimensionality), and I’m pretty sure there are dimensions corresponding to things like “if 1st character an ‘a’, 1st is ‘b’, 2nd is ‘a’ etc. So tokens aren’t as important.
- kadushka 1y agoIs there any evidence to support your hypothesis?
- boroboro4 1y agoGood question! I did a small experiment: trained a small logistics regression from embedding vectors into 1st/2nd/3rd character in token: https://chatgpt.com/share/6871061a-7948-8007-ab53-5b0b697e902a https://chatgpt.com/share/6871061a-7948-8007-ab53-5b0b697e90... I got 0.863 (for 1st)/0.559 (for 2nd)/0.447 (for 3rd) accuracy for Qwen 3 8B model embeddings. Note the code is hacky and might be wrong in ways + in reality transformers do know more because here I utilize only embedding layer. However it does show there are very clear signals on characters in tokens in embedding vectors.
- kadushka 1y agoThank you! I guess if there's enough spelling related text in the dataset, a model is forced to learn some info about token composition in order to predict such texts. I wonder if it would help to explicitly insert this info into an embedding vector, similar to how we encode word position info. For example, allocate the first 20 vector elements to represent ASCII codes of token's characters (in some normalized way).
- boroboro4 1y agoOne way here is to use one hot encoding in first (token length * alphabet length) dimensions. But to be frank I don’t think it’s really needed, I bet everything really needed model learns by itself. If I had time I would’ve tried it though :) Bonus content, accuracies for other models (notice DeepSeek!): - Qwen3-32B: 0.873 / 0.585 / 0.467 - Qwen3-235B-A22B: 0.857 / 0.607 / 0.502 - DeepSeek-V3: 0.869 / 0.738 / 0.624