Tokenization inefficiency imposes structural disadvantages on morphologically complex, low-resource languages, inflating compute resources and depressing accuracy. We evaluate 10 large language models (LLMs) on AfriMMLU (9,000 MCQA items; 5 subjects; 16 African languages) and show that fertility (tokens/word) reliably predicts accuracy. Higher fertility consistently predicts lower accuracy across all models and subjects. We further find that reasoning models (DeepSeek, o1) consistently outperform non-reasoning peers across high and low resource languages in the AfriMMLU dataset, narrowing accuracy gaps observed in prior generations. Finally, translating token inflation to economics, a doubling in tokens results in quadrupled training cost and time, underscoring the token tax faced by many languages. These results motivate morphologically aware tokenization, fair pricing, and multilingual benchmarks for equitable natural language processing (NLP).

Many-shot jailbreaking (MSJ) is an adversarial technique that exploits the long context windows of modern LLMs to circumvent model safety training by including in the prompt many examples of a "fake" assistant responding inappropriately before the final request. With enough examples, the model's in-context learning abilities override its safety training, and it responds as if it were the "fake" assistant. In this work, we probe the effectiveness of different fine-tuning and input sanitization approaches on mitigating MSJ attacks, alone and in combination. We find incremental mitigation effectiveness for each, and show that the combined techniques significantly reduce the effectiveness of MSJ attacks, while retaining model performance in benign in-context learning and conversational tasks. We suggest that our approach could meaningfully ameliorate this vulnerability if incorporated into model safety post-training.

Activation engineering is becoming increasingly popular as a means of online control of large language models (LLMs). In this work, I extend the idea of active steering with vectors that represent a behavioral direction of interest to tuning those vectors directly into the model, obviating the need for online control. First, I identify activation vectors related to honesty in an open-source LLM (Llama- 2-13b-chat). Next, I demonstrate that model output can be made more or less honest by adding positive or negative multiples of these vectors to residual stream activations during generation. Then, I show that a similar effect can be achieved by fine-tuning the vectors directly into the model, by use of a dual loss function based on the cosine similarity of residual stream activations to the vectors combined with a standard token-based loss ("representation tuning"). Finally, I compare the generations in response to honesty-probing prompts from the resulting models to those from models fine-tuned with a token-based loss alone, and to those from the untuned model subjected to online steering. Overall, fine-tuning the vectors into the models using the cosine similarity plus token loss showed a stronger effect than online steering, and generalized better than using the standard loss, suggesting the potential utility of this approach as a safety measure. Code and data are available at https://github.com/cma1114/representation_tuning; tuned models are available at https://huggingface.co/collections/cackerman/ representation-tuning-66da1e5ab41cd1b824687d9f.