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. 30551
    Prev
    Next
    INDEX
    Explanations

    terms related to sin and guilt

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

    sin and crimes

    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
     indisponible
    -0.55
    olak
    -0.53
    uttavia
    -0.53
    zeera
    -0.52
     Konkurrenz
    -0.52
     Bühne
    -0.51
    makeConstraints
    -0.51
     bezpośred
    -0.50
     presto
    -0.49
     Locator
    -0.49
    POSITIVE LOGITS
     sins
    0.62
     crimes
    0.59
     pecado
    0.56
     pecados
    0.56
     offenses
    0.54
     crime
    0.54
     sin
    0.53
     sinned
    0.52
     offences
    0.52
     sinful
    0.52
    Activations Density 0.007%

    No Known Activations