7MIT Technology Review — AI·
Anthropic develops Jacobian lens technique to observe internal reasoning in Claude
Anthropic researchers developed a technique called the 'Jacobian lens' that provides visibility into the internal computations of large language models as they process queries and tasks. The tool reportedly reveals a hidden representational space where models appear to work through concepts before generating outputs. MIT Technology Review describes the findings as ranging from mundane to unnerving, suggesting the technique surfaces unexpected model behaviors or internal states.
Related guides (4)
Related events (8)
Anthropic introduces extended thinking mode in Claude 3.7 Sonnet with visible reasoning traces
Anthropic released Claude 3.7 Sonnet with an 'extended thinking' capability that allows the model to allocate more compute and reasoning time to difficult problems, with a configurable 'thinking budget' for developers. The model's internal reasoning chain is exposed in raw form as a research preview, enabling transparency but raising faithfulness concerns — Anthropic notes that models often make decisions based on factors not explicitly discussed in their visible thinking. The release also includes improved agentic capabilities ('action scaling') demonstrated on OSWorld computer-use benchmarks and a Pokémon Red gameplay evaluation.
Anthropic demonstrates feature steering in Claude 3 Sonnet via interpretability research
Anthropic released a 24-hour public demo called 'Golden Gate Claude' to illustrate findings from a major interpretability paper on Claude 3 Sonnet. The research identifies millions of internal 'features' — neuron combinations that activate for specific concepts — and shows these can be surgically amplified or suppressed to alter model behavior without prompting or fine-tuning. The Golden Gate Bridge feature was amplified as a demonstration, causing the model to reference the bridge in nearly all responses. Anthropic argues this mechanistic control over internal activations has direct implications for AI safety, including the ability to modulate safety-relevant features like those tied to deception or dangerous code.
Claude 3.7 Sonnet and Claude Code: Anthropic's First Hybrid Reasoning Model and Agentic Coding Tool
Anthropic has released Claude 3.7 Sonnet, described as their most capable model to date and the first hybrid reasoning model on the market, capable of operating in both standard and extended thinking modes within a single unified model. The model achieves state-of-the-art results on SWE-bench Verified and TAU-bench, with particular strength in coding and front-end web development. Alongside the model, Anthropic is launching Claude Code in limited research preview, a command-line agentic coding tool that can read/edit files, run tests, and push to GitHub. Pricing remains unchanged at $3/M input and $15/M output tokens, with availability across Claude.ai plans, Amazon Bedrock, and Google Cloud Vertex AI.
ReasoningLens: Open-source framework for hierarchical visualization and auditing of LLM reasoning chains
ReasoningLens is an open-source framework for visualizing and diagnostically auditing the long chain-of-thought traces produced by large reasoning models. It structures traces into interactive hierarchies separating high-level strategy from low-level execution, uses an agentic auditor for automated error detection, and synthesizes model-specific reasoning profiles to surface blind spots. The work targets a growing transparency problem as reasoning models produce increasingly long and opaque inference traces.
Anthropic partners with U.S. National Labs for 1,000 Scientist AI Jam evaluating Claude on scientific tasks
Anthropic is participating in the U.S. Department of Energy's first 1,000 Scientist AI Jam, bringing together scientists across multiple National Laboratories to evaluate frontier AI models on scientific research and national security applications. Claude 3.7 Sonnet, recently launched as the first hybrid reasoning model, will be a primary subject of evaluation across tasks including hypothesis generation, experiment planning, code generation, and result analysis. This builds on Anthropic's April 2024 collaboration with the National Nuclear Security Administration, which was the first instance of a frontier lab evaluating a model in a Top Secret classified environment. The partnership signals deepening government-industry collaboration on AI for scientific discovery and national security.
Anthropic publishes large-scale study of how university students use Claude
Anthropic analyzed one million anonymized student conversations on Claude.ai to produce one of the first large-scale empirical studies of real-world AI usage in higher education. Key findings: Computer Science students are heavily overrepresented (36.8% of conversations vs. 5.4% of U.S. degrees), while Business, Health, and Humanities students underuse the tool relative to enrollment. Students primarily engage in higher-order cognitive tasks per Bloom's Taxonomy—creating and analyzing—though the study raises concerns about offloading critical thinking. The analysis used Anthropic's internal Clio tool, which aggregates conversation patterns while stripping personal information.
We Got Claude to Fine-Tune an Open Source LLM
Hugging Face demonstrates using Claude (Anthropic's model) as an orchestrating agent to autonomously fine-tune an open-source LLM, showcasing an agentic workflow for model training. The post illustrates how a frontier model can handle the end-to-end process of dataset preparation, training configuration, and execution for a smaller open-weights model. This represents a practical example of AI-assisted ML engineering and agent-tool ecosystem development.
How scientists are using Claude to accelerate research and discovery
Anthropic describes how researchers are deploying Claude-powered systems across scientific workflows, highlighting three case studies: Biomni (a Stanford agentic platform integrating hundreds of biomedical tools), the Cheeseman Lab (automating large-scale gene knockout experiment interpretation), and others. The piece details Claude for Life Sciences and the AI for Science program, which provides free API credits to high-impact research projects. Specific benchmarks cited include compressing months-long GWAS analyses to 20 minutes and analyzing 336,000 single-cell datasets to identify novel transcription factors.