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. 18-GEMMASCOPE-RES-16K
    6. 3851
    Prev
    Next
    INDEX
    Explanations

    references to legal terms and proceedings

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

    Follows pronouns

    np_acts-logits-general · gemini-2.0-flash

    this We that her

    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
    <bos>
    -1.02
    kos
    -0.36
    jo
    -0.36
    ρυ
    -0.36
    Ã
    -0.35
    </tr>
    -0.35
    úb
    -0.35
    нен
    -0.35
    ind
    -0.34
     queſta
    -0.34
    POSITIVE LOGITS
    ^(@)
    0.44
    drawal
    0.37
    >\<^
    0.35
     [],
    
    0.34
    csolódó
    0.34
     Theſe
    0.33
    pecially
    0.32
    ViewImports
    0.32
    ideration
    0.31
    ^(@
    0.31
    Activations Density 39.729%

    No Known Activations