© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 11953
    Prev
    Next
    INDEX
    Explanations

    I am unable to determine a concise summary of what the neuron is looking for. From what I can see it isn't keying off of particular words, but is seemingly instead keying off of sentence structure, like questions ending in ?, or exclamations ending in !

    oai_token-act-pair · gemini-2.0-flash

    punctuation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     highly
    -1.03
    Highly
    -0.93
    highly
    -0.90
     Highly
    -0.88
     altamente
    -0.65
     высоко
    -0.47
    asticity
    -0.43
     grund
    -0.43
     високо
    -0.42
    !
    -0.41
    POSITIVE LOGITS
    endphp
    0.67
     NavController
    0.61
    glEnd
    0.59
     wireType
    0.59
     Can
    0.57
    jmniej
    0.56
    databinding
    0.56
    ifty
    0.56
    CascadeType
    0.55
     Won
    0.55
    Activations Density 1.662%

    No Known Activations