© 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. 38-GEMMASCOPE-RES-16K
    11. 11705
    Prev
    Next
    INDEX
    Explanations

    phrases that evaluate performance or quality, particularly in a context of feedback or assessment

    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
     unbedingt
    -0.47
     настолько
    -0.40
    niająca
    -0.39
     profundamente
    -0.38
     невозможно
    -0.37
     poffible
    -0.36
     súper
    -0.36
     tellement
    -0.36
    究極
    -0.36
    nocześnie
    -0.35
    POSITIVE LOGITS
     decent
    2.03
     decently
    1.82
     reasonably
    1.79
     fairly
    1.73
     Decent
    1.60
     reasonable
    1.59
    Decent
    1.59
    reasonably
    1.50
     respectable
    1.46
     moderately
    1.38
    Activations Density 0.818%

    No Known Activations