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. 10-GEMMASCOPE-RES-16K
    6. 2998
    Prev
    Next
    INDEX
    Explanations

    terms related to temporary conditions or actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     réforme
    -0.54
    #+#
    -0.54
     réseau
    -0.52
    tich
    -0.52
    penet
    -0.51
     intégrée
    -0.50
     condominio
    -0.50
    atte
    -0.48
    ellidos
    -0.48
    Kast
    -0.47
    POSITIVE LOGITS
     temporary
    1.50
     Temporary
    1.31
    temporary
    1.30
     temporarily
    1.29
    Temporary
    1.27
     tijde
    1.16
     temporaire
    1.16
     temporales
    1.12
    PORARY
    1.12
     tempor
    1.10
    Activations Density 0.265%

    No Known Activations