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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 35121
    Prev
    Next
    INDEX
    Explanations

    words or tokens that start with an underscore

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    count, id, entries, switches, iter, width, notifications, actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ">—
    -0.34
    ="?
    -0.29
    éndole
    -0.29
     ++)
    -0.29
     mergeFrom
    -0.28
    findpost
    -0.28
     including
    -0.27
    ientí
    -0.27
    )$-
    -0.27
    enterOuterAlt
    -0.27
    POSITIVE LOGITS
     esternos
    0.59
     BoxFit
    0.57
    IPAC
    0.56
     Bins
    0.55
    ########.
    0.55
    ᅝ
    0.54
    ंदीखरीदारी
    0.53
    
    0.53
     Мексичка
    0.53
    ᠁
    0.52
    Activations Density 0.001%

    No Known Activations