© 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-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 16050
    Prev
    Next
    INDEX
    Explanations

    phrases indicating conditional statements or scenarios involving the subject "we."

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

    that indicate an assumption or choice

    np_acts-logits-general · gemini-2.0-flash

    conditional assumption or action

    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
     Eindruck
    -0.47
     quedar
    -0.46
     shares
    -0.45
     Catch
    -0.45
     ajudar
    -0.45
     допо
    -0.45
    Pont
    -0.44
     rév
    -0.44
     helping
    -0.43
    auro
    -0.43
    POSITIVE LOGITS
     assume
    0.80
     consider
    0.75
     regard
    0.66
     treat
    0.66
     imagine
    0.65
     assumes
    0.65
    ymce
    0.65
    treat
    0.65
     NSCoder
    0.63
     betrachtet
    0.63
    Activations Density 1.260%

    No Known Activations