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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 65378
    Prev
    Next
    INDEX
    Explanations

    legal terminology and concepts related to the presumption of correctness and burden of proof in judicial determinations

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

    presumed, presumption

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MessageTagHelper
    -0.41
    ArrowToggle
    -0.41
     AspNetCore
    -0.40
    MessageOf
    -0.39
    ScopeManager
    -0.39
     utafitiHapana
    -0.38
     ExecuteAsync
    -0.37
    dotti
    -0.36
     nép
    -0.35
     besides
    -0.35
    POSITIVE LOGITS
     presumption
    1.66
     presumed
    1.60
     presume
    1.52
     assumed
    1.38
     assume
    1.32
     presump
    1.28
    pres
    1.25
     Pres
    1.23
     pres
    1.23
    assumed
    1.22
    Activations Density 0.244%

    No Known Activations