Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 111595
    Prev
    Next
    INDEX
    Explanations

    programming keywords and concepts related to conditional logic and data structures

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

    fullWidth or horizontal

    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
     defaultstate
    -0.44
    ところで
    -0.37
    +:+
    -0.34
    enterOuterAlt
    -0.34
     excellently
    -0.33
     pôr
    -0.32
     діа
    -0.32
     altrett
    -0.32
     laid
    -0.32
    勿論
    -0.32
    POSITIVE LOGITS
     հղումներ
    0.56
    исленность
    0.56
    WithIOException
    0.53
     Taktlose
    0.51
    vician
    0.49
     ſche
    0.49
     Potosí
    0.49
    contentLoaded
    0.46
     yyr
    0.46
     fubject
    0.46
    Activations Density 0.060%

    No Known Activations