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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 3678
    Prev
    Next
    INDEX
    Explanations

    phrases referencing actions or causality involving the word "by."

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

    Preposition "by" preceding a verb

    np_acts-logits-general · gemini-2.0-flash

    explaining actions with by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     recovers
    -0.69
     assigns
    -0.66
     снять
    -0.65
     domestiques
    -0.65
     habet
    -0.64
    送到
    -0.64
    polate
    -0.64
    allocate
    -0.64
     executes
    -0.63
    Remove
    -0.63
    POSITIVE LOGITS
     being
    0.81
     CreateTagHelper
    0.69
    ValueStyle
    0.61
     becoming
    0.60
    principalColumn
    0.60
     Administrativna
    0.60
     étant
    0.59
     also
    0.59
     fiind
    0.58
     gynhyrchwyd
    0.57
    Activations Density 0.336%

    No Known Activations