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. 41-GEMMASCOPE-RES-16K
    6. 7243
    Prev
    Next
    INDEX
    Explanations

    phrases related to account verification and management

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.41.hook_resid_post
    Hook Layer
    41
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Tujuan
    -0.32
    NUMX
    -0.32
     Masyarakat
    -0.29
     quedado
    -0.29
     Ferner
    -0.29
     Hivatalos
    -0.28
    ksanakan
    -0.28
    sie
    -0.28
     Cinta
    -0.28
     Polri
    -0.27
    POSITIVE LOGITS
     něco
    0.61
     nějak
    0.60
     ten
    0.59
     taky
    0.59
     صوتيه
    0.57
     kafe
    0.57
     teda
    0.57
     když
    0.57
     pará
    0.56
     co
    0.56
    Activations Density 0.027%

    No Known Activations