© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 28-GEMMASCOPE-RES-16K
    11. 11927
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to clinical studies and health-related data

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    preceding company or study names

    np_acts-logits-general · gemini-2.0-flash

    unique positive descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     crappy
    -0.74
     fucked
    -0.73
     shitty
    -0.72
     stuff
    -0.69
     fucking
    -0.68
     thingy
    -0.67
    AutoScaleMode
    -0.65
     FUCKING
    -0.65
     iirc
    -0.62
     stupid
    -0.61
    POSITIVE LOGITS
     uniquely
    0.51
     underscored
    0.49
     underscores
    0.48
     captivating
    0.47
    wavering
    0.47
     revital
    0.46
     unwavering
    0.46
     visionary
    0.45
    Напомним
    0.45
     remer
    0.45
    Activations Density 0.822%

    No Known Activations