© 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. 14397
    Prev
    Next
    INDEX
    Explanations

    the word "present" or phrases that introduce a counterpoint

    oai_token-act-pair · gemini-2.0-flash

    other present

    np_max-act · gemini-2.0-flash

    say "hand" Method used: 2 — because tokens after are often "hand" — say "hand"

    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
     myſelf
    -0.75
    batore
    -0.67
    medriver
    -0.66
     Monfieur
    -0.65
    ardized
    -0.65
     raiſ
    -0.65
    PerformLayout
    -0.65
    AddWithValue
    -0.64
     ſy
    -0.64
     perſon
    -0.63
    POSITIVE LOGITS
     criticised
    0.59
     Programmes
    0.57
    COUVER
    0.56
     недавно
    0.55
     идёт
    0.55
    NameInMap
    0.53
    الحياه
    0.51
     honom
    0.50
     centre
    0.50
    شهاد
    0.50
    Activations Density 1.599%

    No Known Activations