© 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
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 10012
    Prev
    Next
    INDEX
    Explanations

    key elements related to the effectiveness and emotional impact of horror films

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    seamnă
    -0.35
     horizontal
    -0.33
    Népesség
    -0.33
    elser
    -0.32
     Auss
    -0.32
    ‌آم
    -0.31
     considérons
    -0.30
     kasarigan
    -0.30
    sequelize
    -0.29
     listo
    -0.29
    POSITIVE LOGITS
     réuss
    0.73
     succeeds
    0.73
     manages
    0.71
     succeed
    0.70
     lyck
    0.66
     suffers
    0.63
     treads
    0.62
     réussi
    0.62
     réussir
    0.62
     manage
    0.60
    Activations Density 0.305%

    No Known Activations