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. 523
    Prev
    Next
    INDEX
    Explanations

    detailed descriptions of effort and craftsmanship in creative processes

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

    [noun] went into

    np_acts-logits-general · gemini-2.0-flash

    thinking behind what went into

    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
     TextInputType
    -0.67
    URLException
    -0.56
     kasarigan
    -0.55
    medies
    -0.53
     Paglinawan
    -0.53
    TagMode
    -0.53
    RTEX
    -0.52
     Мексичка
    -0.51
    leby
    -0.51
     JAXBElement
    -0.51
    POSITIVE LOGITS
     behind
    1.38
    behind
    1.37
    Behind
    1.35
     dietro
    1.28
     detrás
    1.25
     Behind
    1.24
     BEHIND
    1.16
     derrière
    1.00
     bakom
    0.99
     trás
    0.82
    Activations Density 0.158%

    No Known Activations