© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 28-GEMMASCOPE-RES-16K
    11. 5560
    Prev
    Next
    INDEX
    Explanations

    instances of life-threatening scenarios and their consequences

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

    about survival or safety

    np_acts-logits-general · gemini-2.0-flash

    survived, safe, or unharmed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    __(/*!
    -0.33
     fili
    -0.31
     restricting
    -0.30
    はず
    -0.30
    nesc
    -0.29
     Unlimited
    -0.29
     unlimited
    -0.29
     брать
    -0.29
     looping
    -0.29
     Wide
    -0.28
    POSITIVE LOGITS
     safely
    1.45
     survived
    1.41
     safe
    1.41
     survive
    1.36
     unharmed
    1.34
     survives
    1.32
    safe
    1.20
    survi
    1.18
     Safely
    1.15
     surviving
    1.13
    Activations Density 0.282%

    No Known Activations