Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15866
    Prev
    Next
    INDEX
    Explanations

    names of professional motorsport drivers and references to their careers

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

    racing and motorsports

    np_acts-logits-general · gemini-2.0-flash

    racing drivers and cars

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Taktlose
    -0.60
     rústico
    -0.54
    oneofs
    -0.53
    évaluateur
    -0.52
     informée
    -0.48
     MainAxisSize
    -0.48
    enfance
    -0.47
    ništ
    -0.47
     resourceCulture
    -0.46
    bufio
    -0.46
    POSITIVE LOGITS
     racing
    0.84
     race
    0.81
     Racing
    0.71
     motorsport
    0.70
     racers
    0.68
    racing
    0.68
     racer
    0.67
     Race
    0.67
     raced
    0.66
    Racing
    0.66
    Activations Density 0.142%

    No Known Activations