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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 466
    Prev
    Next
    INDEX
    Explanations

    a focus on entities, specifically relationships and actions involving defendants in legal contexts

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

    Follows a verb in the past tense

    np_acts-logits-general · gemini-2.0-flash

    past tense verbs after subjects

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    دانشنامهٔ
    -0.81
     Normdatei
    -0.77
     Infórmanos
    -0.71
    styleType
    -0.69
     invokingState
    -0.68
    LEncoder
    -0.68
    parsedMessage
    -0.68
    Архівовано
    -0.65
    __*/
    -0.64
     cherchés
    -0.63
    POSITIVE LOGITS
     likewise
    0.37
     theirs
    0.36
     correspondingly
    0.33
     himself
    0.33
     too
    0.32
     peores
    0.32
     perfiles
    0.32
    便
    0.30
     aparcamiento
    0.30
     pruebas
    0.29
    Activations Density 0.672%

    No Known Activations