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-2-9B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 1833
    Prev
    Next
    INDEX
    Explanations

    various forms and contexts of the word "force."

    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
     Ambiental
    -0.55
     Atlántico
    -0.49
     Sábado
    -0.47
    hamcrest
    -0.45
    MethodManager
    -0.45
     rueda
    -0.45
     Siglo
    -0.45
     Administração
    -0.44
    viewWillAppear
    -0.44
     OnInit
    -0.43
    POSITIVE LOGITS
    force
    1.52
     force
    1.47
     Force
    1.44
     FORCE
    1.35
    Force
    1.35
    FORCE
    1.22
    forces
    1.13
     Forces
    1.09
     forces
    1.09
    Forces
    1.02
    Activations Density 0.112%

    No Known Activations