© 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
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 8250
    Prev
    Next
    INDEX
    Explanations

    the word "How" or "how"

    oai_token-act-pair · gemini-2.0-flash

    questions starting with 'How'

    np_token-act-pair-logits · gpt-4o-mini

    how

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
    期刊论文
    -0.74
    MessageTagHelper
    -0.73
    OGND
    -0.70
     виправивши
    -0.70
     OFDb
    -0.69
     AssemblyCulture
    -0.62
    ModelSerializer
    -0.61
    Naissance
    -0.61
    PRNewswire
    -0.61
    PyExc
    -0.60
    POSITIVE LOGITS
     long
    0.57
     old
    0.57
     often
    0.53
    ecer
    0.48
    long
    0.46
    tos
    0.45
    AppMethodBeat
    0.44
     Long
    0.44
     about
    0.43
    ])).
    0.43
    Activations Density 0.317%

    No Known Activations