© 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. 14574
    Prev
    Next
    INDEX
    Explanations

    references to decisions, choices, or parameters in various contexts

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

    substitutions or alternatives

    np_acts-logits-general · gemini-2.0-flash

    rather than

    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
     propOrder
    -0.65
    ⏜
    -0.59
    ſſung
    -0.56
    brigens
    -0.55
     zuſammen
    -0.54
     geweſen
    -0.53
    ElementException
    -0.53
     Exacts
    -0.52
     pleaſure
    -0.52
    <pad>
    -0.52
    POSITIVE LOGITS
     rather
    3.94
     instead
    3.83
    instead
    3.42
    rather
    3.36
     plutôt
    3.16
     Rather
    3.13
    Rather
    3.08
    Instead
    3.06
     Instead
    3.03
     zamiast
    2.98
    Activations Density 2.739%

    No Known Activations