© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 9485
    Prev
    Next
    INDEX
    Explanations

    discourse markers and phrases indicating uncertainty or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    or to

    np_acts-logits-general · gemini-2.5-flash-lite

    the presence of common function/discourse words (auxiliaries, pronouns, connectives and adverbial markers) indicating sentence-level grammatical scaffolding.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ⟬
    -0.84
    StoryboardSegue
    -0.82
    uxxxx
    -0.78
     незавершена
    -0.78
    featureID
    -0.77
     мәкал
    -0.73
     Infórmanos
    -0.73
     Мексичка
    -0.72
    enterOuterAlt
    -0.68
     kaarangay
    -0.67
    POSITIVE LOGITS
     Bernstein
    0.43
     and
    0.40
    anne
    0.37
    【
    0.35
    table
    0.35
    bers
    0.34
     foto
    0.34
    iente
    0.34
    lask
    0.33
     Newman
    0.33
    Activations Density 0.828%

    No Known Activations