© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 30-GEMMASCOPE-RES-16K
    11. 8890
    Prev
    Next
    INDEX
    Explanations

    phrases related to perceived threats or dangers

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

    "threat" or "danger"

    np_acts-logits-general · gemini-2.0-flash

    threats and danger

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ErrIntOverflow
    -0.34
    GOTREF
    -0.32
     št
    -0.30
    凑
    -0.30
     Möbel
    -0.30
    netics
    -0.29
     yardımcı
    -0.28
     RIPRODUZIONE
    -0.27
     Rauch
    -0.26
     fatigued
    -0.26
    POSITIVE LOGITS
     threat
    2.55
     threats
    2.33
    threat
    2.27
     Threat
    2.08
    Threat
    2.02
     danger
    1.96
     threaten
    1.95
     Threats
    1.92
    Threats
    1.86
     threatening
    1.84
    Activations Density 0.475%

    No Known Activations