© 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. 15529
    Prev
    Next
    INDEX
    Explanations

    mentions of the earth and land, but also "wrong" and somewhat related terms.

    oai_token-act-pair · gemini-2.0-flash

    wrongness

    np_max-act · gemini-2.0-flash

    wrong Method used: 1 — because many top activating tokens are "wrong"

    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
    ⓧ
    -1.07
     صوتيه
    -1.02
    StoryboardSegue
    -1.01
    SequentialGroup
    -0.97
     nakalista
    -0.93
    contentLoaded
    -0.90
    SharedCtor
    -0.90
    AutoField
    -0.88
     defaultstate
    -0.88
    principalColumn
    -0.86
    POSITIVE LOGITS
    2
    0.48
     
    0.46
    甚至
    0.46
     contrário
    0.45
     skates
    0.44
     stings
    0.44
     -
    0.44
     брен
    0.43
    Closer
    0.42
    even
    0.42
    Activations Density 1.822%

    No Known Activations