© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 4495
    Prev
    Next
    INDEX
    Explanations

    the word "separate", and sometimes words near it

    oai_token-act-pair · gemini-2.0-flash

    separate

    np_max-act · gemini-2.0-flash

    ▁separate

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
    ↵↵
    -1.09
     (
    -0.86
     
    -0.80
     "
    -0.78
     and
    -0.78
     I
    -0.75
     in
    -0.73
    ↵
    -0.72
     A
    -0.71
     D
    -0.71
    POSITIVE LOGITS
     Monfieur
    1.33
     Theſe
    1.20
     Reſ
    1.13
     Houſe
    1.11
     Cæsar
    1.09
     Verſ
    1.08
     Efq
    1.08
     avoient
    1.06
     Diſ
    1.05
     itſelf
    1.05
    Activations Density 2.945%

    No Known Activations