© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 5283
    Prev
    Next
    INDEX
    Explanations

    materials and their properties in a scientific context, particularly related to medical applications

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    This neuron detects mentions of specific materials or substances (and their descriptors) in technical or scientific contexts.

    oai_token-act-pair · o4-miniTriggered by @phamvanluong2003

    Mentions of materials, material types, or material-related technical properties (e.g., specific materials, scaffolds, implants, and engineering/biomedical material terms).

    oai_token-act-pair · gpt-5-miniTriggered by @phamvanluong2003
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1385
    +0.20
    0.7%
    876
    +0.17
    0.5%
    736
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    736
    +0.20
    0.07
    1385
    +0.17
    0.04
    394
    +0.10
    0.04
    Negative Logits
     panik
    -1.33
     praktik
    -1.31
     akut
    -1.30
     kosme
    -1.28
     minimalis
    -1.24
     protokol
    -1.23
     keramik
    -1.23
     radikal
    -1.22
     Muhamma
    -1.22
     alkoh
    -1.22
    POSITIVE LOGITS
     material
    0.65
     automne
    0.65
     materials
    0.63
     thickness
    0.60
     polymers
    0.60
    Bahan
    0.60
     durable
    0.58
     polymer
    0.58
     lightweight
    0.57
     poroz
    0.57
    Activations Density 0.468%

    No Known Activations