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. 20-GEMMASCOPE-RES-16K
    6. 1418
    Prev
    Next
    INDEX
    Explanations

    references to multi-dimensional or multi-faceted concepts, especially in scientific contexts

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

    starting with "multi"

    np_acts-logits-general · gemini-2.0-flash

    multi- followed by another prefix

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     للمعارف
    -0.83
    eseorang
    -0.65
    OrNil
    -0.65
    #+#
    -0.65
    }');
    -0.63
    )))),
    -0.62
    OGND
    -0.60
    $',
    -0.59
    XmlEnum
    -0.58
    PerformLayout
    -0.58
    POSITIVE LOGITS
     Multi
    1.04
    Multi
    1.01
     multi
    1.00
    multi
    0.94
     MULTI
    0.92
    MULTI
    0.87
     MULT
    0.86
    MULT
    0.75
     multifunctional
    0.64
    iedenen
    0.59
    Activations Density 0.068%

    No Known Activations