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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 1727
    Prev
    Next
    INDEX
    Explanations

    specific elements and variations in structured contexts or categories

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    GEBURTSDATUM
    -0.84
    principalColumn
    -0.80
    Personendaten
    -0.66
     مشين
    -0.63
    OGND
    -0.59
    PasswordEncoder
    -0.59
    ThroughAttribute
    -0.59
     contextLoads
    -0.59
     Administrativna
    -0.59
     jadx
    -0.58
    POSITIVE LOGITS
     different
    0.49
    Different
    0.47
     Different
    0.44
     poszczegól
    0.42
    different
    0.41
     разные
    0.41
     verschillende
    0.40
     unterschied
    0.39
     különböző
    0.39
     verschiedenen
    0.38
    Activations Density 0.838%

    No Known Activations