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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 2768
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names, particularly related to chameleons and associated characters or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /*
    -1.10
    >*/
    -0.89
     jScrollPane
    -0.87
     disambiguazione
    -0.82
    /**
    -0.81
    )*/
    -0.78
    kegaard
    -0.78
    stalt
    -0.76
     Isaacs
    -0.74
    UPI
    -0.73
    POSITIVE LOGITS
     CH
    0.81
     Ch
    0.77
    Ch
    0.77
     Chávez
    0.75
     Shum
    0.73
     Chio
    0.73
     Chor
    0.72
     ch
    0.71
     Sh
    0.71
    Chy
    0.69
    Activations Density 0.493%

    No Known Activations