© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 15-GEMMASCOPE-RES-16K
    11. 10605
    Prev
    Next
    INDEX
    Explanations

    specific phrases or references that convey a sense of nostalgia or reflection on the past

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 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
     préféré
    -0.66
     favoritas
    -0.60
     Saltar
    -0.58
     atteinte
    -0.56
     suivie
    -0.56
     buona
    -0.55
     lettura
    -0.55
     autant
    -0.55
     pubblica
    -0.55
     ragioni
    -0.55
    POSITIVE LOGITS
     big
    0.64
     pesky
    0.64
     '{@
    0.63
     виправивши
    0.59
     يتيمه
    0.55
     little
    0.53
     usual
    0.52
     onCreateView
    0.51
    apore
    0.51
     old
    0.51
    Activations Density 0.120%

    No Known Activations