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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14205
    Prev
    Next
    INDEX
    Explanations

    technical specifications and compatibility information related to electronic devices and software

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

    Compatibility with operating systems or software

    np_acts-logits-general · gemini-2.0-flash

    Android Auto

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     myſelf
    -0.54
     Houſe
    -0.51
    UserScript
    -0.50
     Exactos
    -0.49
     houſe
    -0.48
     ſta
    -0.48
     ſon
    -0.48
     interstate
    -0.47
    tiqueta
    -0.47
    parsedMessage
    -0.47
    POSITIVE LOGITS
     cantantes
    0.38
     plupart
    0.37
     compatible
    0.35
     aanbied
    0.35
     suaminya
    0.34
     preghi
    0.34
    spesies
    0.34
     freundlichen
    0.33
     férias
    0.33
     الاطلاع
    0.33
    Activations Density 0.289%

    No Known Activations