© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 3-GEMMASCOPE-RES-16K
    11. 3464
    Prev
    Next
    INDEX
    Explanations

    names that start with "Har" or include variations of it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nodoc
    -0.52
    endregion
    -0.51
     Boi
    -0.46
    󠁴
    -0.45
    adone
    -0.45
     betweenstory
    -0.45
     Carbon
    -0.44
     Theil
    -0.44
    >>()
    -0.44
    monton
    -0.41
    POSITIVE LOGITS
     Har
    3.11
    Har
    3.11
     har
    3.09
     HAR
    2.80
    har
    2.80
    HAR
    2.64
     HARR
    1.97
     هار
    1.97
     Harbour
    1.96
     Harbor
    1.84
    Activations Density 0.129%

    No Known Activations