Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 10408
    Prev
    Next
    INDEX
    Explanations

    phrases indicating conditions of existence or transitions in narratives

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

    only / not until

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pstead
    -0.43
    reportWebVitals
    -0.35
    ogados
    -0.35
    Atsauces
    -0.34
     BoxFit
    -0.34
     كومونز
    -0.34
    rawDesc
    -0.32
    הערות
    -0.32
    ammen
    -0.31
    .*")]
    -0.31
    POSITIVE LOGITS
     only
    2.11
     dopiero
    2.00
    only
    1.69
    Only
    1.67
     Only
    1.66
     seulement
    1.63
     только
    1.56
    只有
    1.48
    ONLY
    1.38
     тільки
    1.35
    Activations Density 0.140%

    No Known Activations