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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28500
    Prev
    Next
    INDEX
    Explanations

    punctuation marks or special characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tartalomajánló
    -0.63
    <eos>
    -0.57
    Gweler
    -0.52
    髦
    -0.51
    fvar
    -0.49
     сочи
    -0.49
    `${
    -0.48
    </tfoot>
    -0.47
    Varint
    -0.47
    izumi
    -0.47
    POSITIVE LOGITS
    ]")]
    0.69
     للمعارف
    0.69
    enumi
    0.66
    \{\\
    0.66
    SequentialGroup
    0.66
     surla
    0.65
    enumii
    0.64
    VersionUID
    0.64
     PhpStorm
    0.63
    Hauptartikel
    0.63
    Activations Density 0.002%

    No Known Activations