© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 3868
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "have" and its various forms indicating possession or existence in context

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

    "has" followed by a past participle

    np_acts-logits-general · gemini-2.0-flash

    has implications or consequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     wrote
    -0.86
     knew
    -0.80
    wrote
    -0.78
     chose
    -0.74
    EndInit
    -0.74
    knew
    -0.69
     Knew
    -0.69
    ThroughAttribute
    -0.66
     undertook
    -0.65
     spoke
    -0.63
    POSITIVE LOGITS
     implications
    0.76
     arisen
    0.65
     ramifications
    0.63
     nothing
    0.60
     consequences
    0.59
     oprot
    0.58
     repercussions
    0.58
     little
    0.57
     been
    0.57
     similarities
    0.56
    Activations Density 0.252%

    No Known Activations