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. 85851
    Prev
    Next
    INDEX
    Explanations

    coding syntax or formatting markers

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

    code/latex section starts

    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
     disambiguazione
    -1.00
    GEBURTS
    -0.88
     Administrativna
    -0.81
     queſta
    -0.79
    AndEndTag
    -0.79
    Personendaten
    -0.77
     surla
    -0.77
     Italijani
    -0.76
    niſſe
    -0.76
    bootstrapcdn
    -0.75
    POSITIVE LOGITS
    ada
    0.31
    frage
    0.30
     Beginning
    0.29
     beginning
    0.28
     begun
    0.28
    query
    0.27
    everything
    0.26
    里
    0.25
    Everything
    0.25
    始
    0.25
    Activations Density 0.001%

    No Known Activations