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. 2-GEMMASCOPE-RES-16K
    6. 13096
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of "bo" or "BO" which may refer to a specific term or abbreviation in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Wicidata
    -0.51
    midler
    -0.48
     flechas
    -0.46
    วด
    -0.44
     regalías
    -0.44
     pluie
    -0.43
    phazard
    -0.43
     uttry
    -0.43
     sanitaires
    -0.43
    jspb
    -0.42
    POSITIVE LOGITS
    bo
    1.16
     bo
    1.03
    BO
    0.91
     BO
    0.80
    Bo
    0.77
    blo
    0.71
    boat
    0.70
    booth
    0.70
     blo
    0.68
     Bo
    0.67
    Activations Density 0.523%

    No Known Activations