© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 12-GEMMASCOPE-RES-16K
    11. 8828
    Prev
    Next
    INDEX
    Explanations

    questions and discussions related to personal preferences and opinions

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

    direct address or imperative statements directed at "you" or "me".

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    second-person direct address—requests, commands, or questions that explicitly involve you/me interaction, often using communication verbs.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Absicht
    -0.38
     dénon
    -0.37
     Überprüfung
    -0.36
    grès
    -0.35
     شك
    -0.35
    Rejected
    -0.34
    申
    -0.34
    Método
    -0.34
     henkil
    -0.34
    Practice
    -0.34
    POSITIVE LOGITS
    sidemargin
    0.78
    mybatisplus
    0.69
    migrationBuilder
    0.69
    AutoScaleMode
    0.69
    AsUp
    0.68
     MainAxisSize
    0.68
     noDo
    0.67
    PMailer
    0.64
    脚注の使い方
    0.63
    таратура
    0.62
    Activations Density 0.469%

    No Known Activations