© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 8951
    Prev
    Next
    INDEX
    Explanations

    terms associated with social gatherings or parties

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

    Events involving hosts, guests, or parties

    np_acts-logits-general · gemini-2.0-flash

    party and guests

    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
     []:
    -0.41
    äk
    -0.39
    ättä
    -0.39
    ment
    -0.38
     délai
    -0.38
    onChanged
    -0.37
     om
    -0.37
    (!__
    -0.36
     метро
    -0.34
    كويكب
    -0.34
    POSITIVE LOGITS
     party
    1.93
    party
    1.77
     PARTY
    1.70
     Party
    1.65
    Party
    1.65
    PARTY
    1.55
     parties
    1.52
     guests
    1.45
     invitations
    1.44
    parties
    1.43
    Activations Density 0.240%

    No Known Activations