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. 80693
    Prev
    Next
    INDEX
    Explanations

    mentions of specific authors and their works, particularly those in dramatic contexts

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

    Literature and film titles

    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
    enumii
    -0.53
     Efq
    -0.50
    UnitTesting
    -0.50
    édrale
    -0.49
    LayoutConstraint
    -0.46
    ExtendWith
    -0.46
    +#+#
    -0.45
    RemoveField
    -0.45
     pstmt
    -0.44
    PMailer
    -0.43
    POSITIVE LOGITS
    Enders
    0.59
     urban
    0.53
    Urban
    0.50
     urbaine
    0.48
     urbano
    0.47
     Urban
    0.47
     tenement
    0.46
     street
    0.44
     slum
    0.41
     urbana
    0.41
    Activations Density 0.036%

    No Known Activations