Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 68864
    Prev
    Next
    INDEX
    Explanations

    phrases relating to visits and encounters in various contexts

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

    visiting, on vacation, first started

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Bewußt
    -0.49
     inconvénients
    -0.43
    ワイイ
    -0.42
     verdaderos
    -0.42
     getItemCount
    -0.40
     deudas
    -0.39
     impreso
    -0.37
     Aufwand
    -0.36
     Fiscalía
    -0.36
     matemáticas
    -0.36
    POSITIVE LOGITS
    enumii
    0.81
    nodoc
    0.64
    IsMutable
    0.63
     Administrativna
    0.63
    脚注の使い方
    0.61
    initState
    0.60
    出版年
    0.60
    UIControlState
    0.59
    makeText
    0.59
    :✨
    0.59
    Activations Density 0.067%

    No Known Activations