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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 62176
    Prev
    Next
    INDEX
    Explanations

    references to authority and organizational structures

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

    Nouns that designate organizational or functional structures such as "authority," "powers," "phase," "formation," "organization," "foundations," "funds," "system," "order," "line," or "relay" - often referring to components that regulate, control, or transmit something within a larger system.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    存于互联网档案馆
    -0.38
    あく
    -0.38
    Rptr
    -0.37
     Rptr
    -0.37
     برانيه
    -0.35
     ͡°
    -0.35
     lautet
    -0.34
    addComponent
    -0.34
    ;"></
    -0.33
    nsec
    -0.32
    POSITIVE LOGITS
    buttonShape
    0.56
    Решение
    0.52
     Longo
    0.48
     Musiker
    0.47
    TextInputLayout
    0.44
    jTable
    0.43
    łko
    0.42
    ciens
    0.42
    cues
    0.42
     rechange
    0.41
    Activations Density 0.296%

    No Known Activations