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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 9858
    Prev
    Next
    INDEX
    Explanations

    references to names, particularly those associated with the term "Kai" and its variations

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

    starting with "Ai", "Kai", "Sai", "Mai", or "Nai"

    np_acts-logits-general · gemini-2.0-flash

    names like kai, ai, sai, mai

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     meantime
    -1.81
     meanwhile
    -1.65
     Meanwhile
    -1.56
    Meanwhile
    -1.53
     frattempo
    -1.33
     mean
    -1.24
    mean
    -1.24
     Mean
    -1.18
    Mean
    -1.18
    Sementara
    -1.03
    POSITIVE LOGITS
    ̯
    1.08
     صوتيه
    0.77
    webElementXpaths
    0.75
    География
    0.70
    ConstraintMaker
    0.70
    ModelState
    0.64
    Kjelder
    0.63
    expiresIn
    0.61
    daos
    0.61
     neuve
    0.60
    Activations Density 0.095%

    No Known Activations