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. 19-GEMMASCOPE-RES-16K
    6. 15570
    Prev
    Next
    INDEX
    Explanations

    questions and calls for confirmation or clarification about specific information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Hook Layer
    19
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     entanto
    -0.86
    ConstraintMaker
    -0.78
     Derfor
    -0.77
    Cependant
    -0.74
    Однако
    -0.73
     obstante
    -0.69
    spesies
    -0.68
    However
    -0.67
    Portail
    -0.67
    however
    -0.65
    POSITIVE LOGITS
    それとも
    0.75
     or
    0.70
     yoksa
    0.59
     Specifically
    0.57
     Seems
    0.56
     Preferably
    0.55
     Or
    0.54
    Seems
    0.53
    Cuz
    0.51
     curiosidad
    0.51
    Activations Density 0.216%

    No Known Activations